conference
stringclasses 6
values | title
stringlengths 8
176
| abstract
stringlengths 228
5k
| decision
stringclasses 9
values |
|---|---|---|---|
ICLR.cc/2020/Conference | A bi-diffusion based layer-wise sampling method for deep learning in large graphs | The Graph Convolutional Network (GCN) and its variants are powerful models for graph representation learning and have recently achieved great success on many graph-based applications. However, most of them target on shallow models (e.g. 2 layers) on relatively small graphs. Very recently, although many acceleration methods have been developed for GCNs training,
it still remains a severe challenge how to scale GCN-like models to larger graphs and deeper layers due to the over-expansion of neighborhoods across layers. In this paper, to address the above challenge, we propose a novel layer-wise sampling strategy, which samples the nodes layer by layer conditionally based on the factors of the bi-directional diffusion between layers. In this way, we potentially restrict the time complexity linear to the number of layers, and construct a mini-batch of nodes with high local bi-directional influence (correlation). Further, we apply the self-attention mechanism to flexibly learn suitable weights for the sampled nodes, which allows the model to be able to incorporate both the first-order and higher-order proximities during a single layer propagation process without extra recursive propagation or skip connection. Extensive experiments on three large benchmark graphs demonstrate the effectiveness and efficiency of the proposed model. | Reject |
ICLR.cc/2021/Conference | ERMAS: Learning Policies Robust to Reality Gaps in Multi-Agent Simulations | Policies for real-world multi-agent problems, such as optimal taxation, can be learned in multi-agent simulations with AI agents that emulate humans. However, simulations can suffer from reality gaps as humans often act suboptimally or optimize for different objectives (i.e., bounded rationality). We introduce $\epsilon$-Robust Multi-Agent Simulation (ERMAS), a robust optimization framework to learn AI policies that are robust to such multi-agent reality gaps. The objective of ERMAS theoretically guarantees robustness to the $\epsilon$-Nash equilibria of other agents – that is, robustness to behavioral deviations with a regret of at most $\epsilon$. ERMAS efficiently solves a first-order approximation of the robustness objective using meta-learning methods. We show that ERMAS yields robust policies for repeated bimatrix games and optimal adaptive taxation in economic simulations, even when baseline notions of robustness are uninformative or intractable. In particular, we show ERMAS can learn tax policies that are robust to changes in agent risk aversion, improving policy objectives (social welfare) by up to 15% in complex spatiotemporal simulations using the AI Economist (Zheng et al., 2020). | Reject |
ICLR.cc/2019/Conference | Recurrent Experience Replay in Distributed Reinforcement Learning | Building on the recent successes of distributed training of RL agents, in this paper we investigate the training of RNN-based RL agents from distributed prioritized experience replay. We study the effects of parameter lag resulting in representational drift and recurrent state staleness and empirically derive an improved training strategy. Using a single network architecture and fixed set of hyper-parameters, the resulting agent, Recurrent Replay Distributed DQN, quadruples the previous state of the art on Atari-57, and matches the state of the art on DMLab-30. It is the first agent to exceed human-level performance in 52 of the 57 Atari games. | Accept (Poster) |
ICLR.cc/2020/Conference | Towards neural networks that provably know when they don't know | It has recently been shown that ReLU networks produce arbitrarily over-confident predictions far away from the
training data. Thus, ReLU networks do not know when they don't know. However, this is a highly important property in safety
critical applications. In the context of out-of-distribution detection (OOD) there have been a number of proposals to mitigate this problem but none of them are able to make any mathematical guarantees. In this paper we propose a new approach to OOD which overcomes both problems. Our approach can be used with ReLU networks and provides provably low confidence predictions far away from the training data as well as the first certificates for low confidence predictions in a neighborhood of an out-distribution point. In the experiments we show that state-of-the-art methods fail in this worst-case setting whereas our model can guarantee its performance while retaining state-of-the-art OOD performance. | Accept (Poster) |
ICLR.cc/2021/Conference | Gradient Based Memory Editing for Task-Free Continual Learning | Prior work on continual learning often operate in a “task-aware” manner, by assuming that the task boundaries and identifies of the data examples are known at all times. While in practice, it is rarely the case that such information are exposed to the methods (i.e., thus called “task-free”)–a setting that is relatively underexplored. Recent attempts on task-free continual learning build on previous memory replay methods and focus on developing memory construction and replay strategies such that model performance over previously seen examples can be best retained. In this paper, looking from a complementary angle, we propose a novel approach to “edit” memory examples so that the edited memory can better retain past performance when they are replayed. We use gradient updates to edit memory examples so that they are more likely to be “forgotten” in the future. Experiments on five benchmark datasets show the proposed method can be seamlessly combined with baselines to significantly improve the performance. | Reject |
ICLR.cc/2023/Conference | Learning from conflicting data with hidden contexts | Classical supervised learning assumes a stable relation between inputs and outputs. However, this assumption is often invalid in real-world scenarios where the input-output relation in the data depends on some hidden contexts. We formulate a more general setting where the training data is sampled from multiple unobservable domains, while different domains may possess semantically distinct input-output maps. Training data exhibits inherent conflict in this setting, rendering vanilla empirical risk minimization problematic. We propose to tackle this problem by introducing an allocation function that learns to allocate conflicting data to different prediction models, resulting in an algorithm that we term LEAF. We draw an intriguing connection between our approach and a variant of the Expectation-Maximization algorithm. We provide theoretical justifications for LEAF on its identifiability, learnability, and generalization error. Empirical results demonstrate the efficacy and potential applications of LEAF in a range of regression and classification tasks on both synthetic data and real-world datasets. | Reject |
ICLR.cc/2019/Conference | Unsupervised Video-to-Video Translation | Unsupervised image-to-image translation is a recently proposed task of translating an image to a different style or domain given only unpaired image examples at training time. In this paper, we formulate a new task of unsupervised video-to-video translation, which poses its own unique challenges. Translating video implies learning not only the appearance of objects and scenes but also realistic motion and transitions between consecutive frames. We investigate the performance of per-frame video-to-video translation using existing image-to-image translation networks, and propose a spatio-temporal 3D translator as an alternative solution to this problem. We evaluate our 3D method on multiple synthetic datasets, such as moving colorized digits, as well as the realistic segmentation-to-video GTA dataset and a new CT-to-MRI volumetric images translation dataset. Our results show that frame-wise translation produces realistic results on a single frame level but underperforms significantly on the scale of the whole video compared to our three-dimensional translation approach, which is better able to learn the complex structure of video and motion and continuity of object appearance. | Reject |
ICLR.cc/2020/Conference | A Dynamic Approach to Accelerate Deep Learning Training | Mixed-precision arithmetic combining both single- and half-precision operands in the same operation have been successfully applied to train deep neural networks. Despite the advantages of mixed-precision arithmetic in terms of reducing the need for key resources like memory bandwidth or register file size, it has a limited capacity for diminishing computing costs and requires 32 bits to represent its output operands. This paper proposes two approaches to replace mixed-precision for half-precision arithmetic during a large portion of the training. The first approach achieves accuracy ratios slightly slower than the state-of-the-art by using half-precision arithmetic during more than 99% of training. The second approach reaches the same accuracy as the state-of-the-art by dynamically switching between half- and mixed-precision arithmetic during training. It uses half-precision during more than 94% of the training process. This paper is the first in demonstrating that half-precision can be used for a very large portion of DNNs training and still reach state-of-the-art accuracy. | Reject |
ICLR.cc/2022/Conference | Trivial or Impossible --- dichotomous data difficulty masks model differences (on ImageNet and beyond) | "The power of a generalization system follows directly from its biases" (Mitchell 1980). Today, CNNs are incredibly powerful generalisation systems---but to what degree have we understood how their inductive bias influences model decisions? We here attempt to disentangle the various aspects that determine how a model decides. In particular, we ask: what makes one model decide differently from another? In a meticulously controlled setting, we find that (1.) irrespective of the network architecture or objective (e.g. self-supervised, semi-supervised, vision transformers, recurrent models) all models end up with a similar decision boundary. (2.) To understand these findings, we analysed model decisions on the ImageNet validation set from epoch to epoch and image by image. We find that the ImageNet validation set, among others, suffers from dichotomous data difficulty (DDD): For the range of investigated models and their accuracies, it is dominated by 46.0% "trivial" and 11.5% "impossible" images (beyond label errors). Only 42.5% of the images could possibly be responsible for the differences between two models' decision boundaries. (3.) Only removing the "impossible" and "trivial" images allows us to see pronounced differences between models. (4.) Humans are highly accurate at predicting which images are "trivial" and "impossible" for CNNs (81.4%). This implies that in future comparisons of brains, machines and behaviour, much may be gained from investigating the decisive role of images and the distribution of their difficulties. | Accept (Poster) |
ICLR.cc/2019/Conference | Tree-Structured Recurrent Switching Linear Dynamical Systems for Multi-Scale Modeling | Many real-world systems studied are governed by complex, nonlinear dynamics. By modeling these dynamics, we can gain insight into how these systems work, make predictions about how they will behave, and develop strategies for controlling them. While there are many methods for modeling nonlinear dynamical systems, existing techniques face a trade off between offering interpretable descriptions and making accurate predictions. Here, we develop a class of models that aims to achieve both simultaneously, smoothly interpolating between simple descriptions and more complex, yet also more accurate models. Our probabilistic model achieves this multi-scale property through of a hierarchy of locally linear dynamics that jointly approximate global nonlinear dynamics. We call it the tree-structured recurrent switching linear dynamical system. To fit this model, we present a fully-Bayesian sampling procedure using Polya-Gamma data augmentation to allow for fast and conjugate Gibbs sampling. Through a variety of synthetic and real examples, we show how these models outperform existing methods in both interpretability and predictive capability. | Accept (Poster) |
ICLR.cc/2022/Conference | Two Birds, One Stone: Achieving both Differential Privacy and Certified Robustness for Pre-trained Classifiers via Input Perturbation | Recent studies have shown that pre-trained classifiers are increasingly powerful to improve the performance on different tasks, e.g, neural language processing, image classification. However, adversarial examples from attackers can trick pre-trained classifiers to misclassify. To solve this challenge, a reconstruction network is built before the public pre-trained classifiers to offer certified robustness and defend against adversarial examples through input perturbation. On the other hand, the reconstruction network requires training on the dataset, which incurs privacy leakage of training data through inference attacks. To prevent this leakage, differential privacy (DP) is applied to offer a provable privacy guarantee on training data through gradient perturbation. Most existing works employ certified robustness and DP independently and fail to exploit the fact that input perturbation designed to achieve certified robustness can achieve (partial) DP. In this paper, we propose perturbation transformation to show how the input perturbation designed for certified robustness can be transformed into gradient perturbation during training. We propose Multivariate Gaussian mechanism to analyze the privacy guarantee of this transformed gradient perturbation and precisely quantify the level of DP achieved by input perturbation. To satisfy the overall DP requirement, we add additional gradient perturbation during training and propose Mixed Multivariate Gaussian Analysis to analyze the privacy guarantee provided by the transformed gradient perturbation and additional gradient perturbation. Moreover, we prove that Mixed Multivariate Gaussian Analysis can work with moments accountant to provide a tight DP estimation. Extensive experiments on benchmark datasets show that our framework significantly outperforms state-of-the-art methods and achieves better accuracy and robustness under the same privacy guarantee. | Reject |
ICLR.cc/2021/Conference | Hybrid Discriminative-Generative Training via Contrastive Learning | Contrastive learning and supervised learning have both seen significant progress and success. However, thus far they have largely been treated as two separate objectives, brought together only by having a shared neural network. In this paper we show that through the perspective of hybrid discriminative-generative training of energy-based models we can make a direct connection between contrastive learning and supervised learning. Beyond presenting this unified view, we show our specific choice of approximation of the energy-based loss significantly improves energy-based models and contrastive learning based methods in confidence-calibration, out-of-distribution detection, adversarial robustness, generative modeling, and image classification tasks. In addition to significantly improved performance, our method also gets rid of SGLD training and does not suffer from training instability. Our evaluations also demonstrate that our method performs better than or on par with state-of-the-art hand-tailored methods in each task. | Reject |
ICLR.cc/2021/Conference | Uncertainty in Neural Processes | We explore the effects of architecture and training objective choice on amortized posterior predictive inference in probabilistic conditional generative models. We aim this work to be a counterpoint to a recent trend in the literature that stresses achieving good samples when the amount of conditioning data is large. We instead focus our attention on the case where the amount of conditioning data is small. We highlight specific architecture and objective choices that we find lead to qualitative and quantitative improvement to posterior inference in this low data regime. Specifically we explore the effects of choices of pooling operator and variational family on posterior quality in neural processes. Superior posterior predictive samples drawn from our novel neural process architectures are demonstrated via image completion/in-painting experiments. | Reject |
ICLR.cc/2023/Conference | Forces are not Enough: Benchmark and Critical Evaluation for Machine Learning Force Fields with Molecular Simulations | Molecular dynamics (MD) simulation techniques are widely used for various natural science applications. Increasingly, machine learning (ML) force field (FF) models begin to replace ab-initio simulations by predicting forces directly from atomic structures. Despite significant progress in this area, such techniques are primarily benchmarked by their force/energy prediction errors, even though the practical use case would be to produce realistic MD trajectories. We aim to fill this gap by introducing a novel benchmark suite for ML MD simulation. We curate representative MD systems, including water, organic molecules, peptide, and materials, and design evaluation metrics corresponding to the scientific objectives of respective systems. We benchmark a collection of state-of-the-art (SOTA) ML FF models and illustrate, in particular, how the commonly benchmarked force accuracy is not well aligned with relevant simulation metrics. We demonstrate when and how selected SOTA methods fail, along with offering directions for further improvement. Specifically, we identify stability as a key metric for ML models to improve. Our benchmark suite comes with a comprehensive open source codebase for training and simulation with ML FFs to facilitate further work. | Reject |
ICLR.cc/2023/Conference | Structural Generalization of Visual Imitation Learning with Position-Invariant Regularization | How the visual imitation learning models can generalize to novel unseen visual observations is a highly challenging problem. Such a generalization ability is very crucial for their real-world applications. Since this generalization problem has many different aspects, we focus on one case called structural generalization, which refers to generalization to unseen task setup, such as a novel setup of object locations in the robotic manipulation problem. In this case, previous works observe that the visual imitation learning models will overfit to the absolute information (e.g., coordinates) rather than the relational information between objects, which is more important for decision making. As a result, the models will perform poorly in novel scenarios. Nevertheless, so far, it remains unclear how we can solve this problem effectively. Our insight into this problem is to explicitly remove the absolute information from the features learned by imitation learning models so that the models can use robust, relational information to make decisions. To this end, we propose a novel, position-invariant regularizer for generalization. The proposed regularizer will penalize the imitation learning model when its features contain absolute, positional information of objects. We carry out experiments on the MAGICAL and ProcGen benchmark, as well as a real-world robot manipulation problem. We find that our regularizer can effectively boost the structural generalization performance of imitation learning models. Through both qualitative and quantitative analysis, we verify that our method does learn robust relational representations.
| Reject |
ICLR.cc/2019/Conference | Learning and Planning with a Semantic Model | Building deep reinforcement learning agents that can generalize and adapt to unseen environments remains a fundamental challenge for AI. This paper describes progresses on this challenge in the context of man-made environments, which are visually diverse but contain intrinsic semantic regularities. We propose a hybrid model-based and model-free approach, LEArning and Planning with Semantics (LEAPS), consisting of a multi-target sub-policy that acts on visual inputs, and a Bayesian model over semantic structures. When placed in an unseen environment, the agent plans with the semantic model to make high-level decisions, proposes the next sub-target for the sub-policy to execute, and updates the semantic model based on new observations. We perform experiments in visual navigation tasks using House3D, a 3D environment that contains diverse human-designed indoor scenes with real-world objects. LEAPS outperforms strong baselines that do not explicitly plan using the semantic content. | Reject |
ICLR.cc/2022/Conference | Learning Vision-Guided Quadrupedal Locomotion End-to-End with Cross-Modal Transformers | We propose to address quadrupedal locomotion tasks using Reinforcement Learning (RL) with a Transformer-based model that learns to combine proprioceptive information and high-dimensional depth sensor inputs. While learning-based locomotion has made great advances using RL, most methods still rely on domain randomization for training blind agents that generalize to challenging terrains. Our key insight is that proprioceptive states only offer contact measurements for immediate reaction, whereas an agent equipped with visual sensory observations can learn to proactively maneuver environments with obstacles and uneven terrain by anticipating changes in the environment many steps ahead. In this paper, we introduce LocoTransformer, an end-to-end RL method that leverages both proprioceptive states and visual observations for locomotion control. We evaluate our method in challenging simulated environments with different obstacles and uneven terrain. We transfer our learned policy from simulation to a real robot by running it indoor and in-the-wild with unseen obstacles and terrain. Our method not only significantly improves over baselines, but also achieves far better generalization performance, especially when transferred to the real robot. Our project page with videos is at https://rchalyang.github.io/LocoTransformer/. | Accept (Spotlight) |
ICLR.cc/2020/Conference | ROS-HPL: Robotic Object Search with Hierarchical Policy Learning and Intrinsic-Extrinsic Modeling | Despite significant progress in Robotic Object Search (ROS) over the recent years with deep reinforcement learning based approaches, the sparsity issue in reward setting as well as the lack of interpretability of the previous ROS approaches leave much to be desired. We present a novel policy learning approach for ROS, based on a hierarchical and interpretable modeling with intrinsic/extrinsic reward setting, to tackle these two challenges. More specifically, we train the low-level policy by deliberating between an action that achieves an immediate sub-goal and the one that is better suited for achieving the final goal. We also introduce a new evaluation metric, namely the extrinsic reward, as a harmonic measure of the object search success rate and the average steps taken. Experiments conducted with multiple settings on the House3D environment validate and show that the intelligent agent, trained with our model, can achieve a better object search performance (higher success rate with lower average steps, measured by SPL: Success weighted by inverse Path Length). In addition, we conduct studies w.r.t. the parameter that controls the weighted overall reward from intrinsic and extrinsic components. The results suggest it is critical to devise a proper trade-off strategy to perform the object search well. | Reject |
ICLR.cc/2018/Conference | i-RevNet: Deep Invertible Networks | It is widely believed that the success of deep convolutional networks is based on progressively discarding uninformative variability about the input with respect to the problem at hand. This is supported empirically by the difficulty of recovering images from their hidden representations, in most commonly used network architectures. In this paper we show via a one-to-one mapping that this loss of information is not a necessary condition to learn representations that generalize well on complicated problems, such as ImageNet. Via a cascade of homeomorphic layers, we build the $i$-RevNet, a network that can be fully inverted up to the final projection onto the classes, i.e. no information is discarded. Building an invertible architecture is difficult, for one, because the local inversion is ill-conditioned, we overcome this by providing an explicit inverse.
An analysis of i-RevNet’s learned representations suggests an alternative explanation for the success of deep networks by a progressive contraction and linear separation with depth. To shed light on the nature of the model learned by the $i$-RevNet we reconstruct linear interpolations between natural image representations. | Accept (Poster) |
ICLR.cc/2023/Conference | Soundness and Completeness: An Algorithmic Perspective on Evaluation of Feature Attribution | Feature attribution is a fundamental approach to explaining neural networks by quantifying the importance of input features for a model's prediction. Although a variety of feature attribution methods have been proposed, there is little consensus on the assessment of attribution methods. In this study, we empirically show the limitations of \emph{order-based} and \emph{model-retraining} metrics. To overcome the limitations and enable evaluation with higher granularity, we propose a novel method to evaluate the \emph{completeness} and \emph{soundness} of feature attribution methods. Our proposed evaluation metrics are mathematically grounded on algorithm theory and require no knowledge of "ground truth" informative features. We validate our proposed metrics by conducting experiments on synthetic and real-world datasets. Lastly, we use the proposed metrics to benchmark a wide range of feature attribution methods. Our evaluation results provide an innovative perspective on comparing feature attribution methods. Code is in the supplementary material. | Reject |
ICLR.cc/2021/Conference | $\alpha$VIL: Learning to Leverage Auxiliary Tasks for Multitask Learning | Multitask Learning is a Machine Learning paradigm that aims to train a range of (usually related) tasks with the help of a shared model. While the goal is often to improve the joint performance of all training tasks, another approach is to focus on the performance of a specific target task, while treating the remaining ones as auxiliary data from which to possibly leverage positive transfer towards the target during training. In such settings, it becomes important to estimate the positive or negative influence auxiliary tasks will have on the target. While many ways have been proposed to estimate task weights before or during training they typically rely on heuristics or extensive search of the weighting space. We propose a novel method called $\alpha$-Variable Importance Learning ($\alpha$VIL) that is able to adjust task weights dynamically during model training, by making direct use of task-specific updates of the underlying model's parameters between training epochs. Experiments indicate that $\alpha$VIL is able to outperform other Multitask Learning approaches in a variety of settings. To our knowledge, this is the first attempt at making direct use of model updates for task weight estimation. | Reject |
ICLR.cc/2018/Conference | Clipping Free Attacks Against Neural Networks | During the last years, a remarkable breakthrough has been made in AI domain
thanks to artificial deep neural networks that achieved a great success in many
machine learning tasks in computer vision, natural language processing, speech
recognition, malware detection and so on. However, they are highly vulnerable
to easily crafted adversarial examples. Many investigations have pointed out this
fact and different approaches have been proposed to generate attacks while adding
a limited perturbation to the original data. The most robust known method so far
is the so called C&W attack [1]. Nonetheless, a countermeasure known as fea-
ture squeezing coupled with ensemble defense showed that most of these attacks
can be destroyed [6]. In this paper, we present a new method we call Centered
Initial Attack (CIA) whose advantage is twofold : first, it insures by construc-
tion the maximum perturbation to be smaller than a threshold fixed beforehand,
without the clipping process that degrades the quality of attacks. Second, it is
robust against recently introduced defenses such as feature squeezing, JPEG en-
coding and even against a voting ensemble of defenses. While its application is
not limited to images, we illustrate this using five of the current best classifiers
on ImageNet dataset among which two are adversarialy retrained on purpose to
be robust against attacks. With a fixed maximum perturbation of only 1.5% on
any pixel, around 80% of attacks (targeted) fool the voting ensemble defense and
nearly 100% when the perturbation is only 6%. While this shows how it is difficult
to defend against CIA attacks, the last section of the paper gives some guidelines
to limit their impact. | Reject |
ICLR.cc/2020/Conference | FSNet: Compression of Deep Convolutional Neural Networks by Filter Summary | We present a novel method of compression of deep Convolutional Neural Networks (CNNs) by weight sharing through a new representation of convolutional filters. The proposed method reduces the number of parameters of each convolutional layer by learning a $1$D vector termed Filter Summary (FS). The convolutional filters are located in FS as overlapping $1$D segments, and nearby filters in FS share weights in their overlapping regions in a natural way. The resultant neural network based on such weight sharing scheme, termed Filter Summary CNNs or FSNet, has a FS in each convolution layer instead of a set of independent filters in the conventional convolution layer. FSNet has the same architecture as that of the baseline CNN to be compressed, and each convolution layer of FSNet has the same number of filters from FS as that of the basline CNN in the forward process. With compelling computational acceleration ratio, the parameter space of FSNet is much smaller than that of the baseline CNN. In addition, FSNet is quantization friendly. FSNet with weight quantization leads to even higher compression ratio without noticeable performance loss. We further propose Differentiable FSNet where the way filters share weights is learned in a differentiable and end-to-end manner. Experiments demonstrate the effectiveness of FSNet in compression of CNNs for computer vision tasks including image classification and object detection, and the effectiveness of DFSNet is evidenced by the task of Neural Architecture Search. | Accept (Poster) |
ICLR.cc/2021/Conference | Deformable Capsules for Object Detection | Capsule networks promise significant benefits over convolutional networks by storing stronger internal representations, and routing information based on the agreement between intermediate representations' projections. Despite this, their success has been mostly limited to small-scale classification datasets due to their computationally expensive nature. Recent studies have partially overcome this burden by locally-constraining the dynamic routing of features with convolutional capsules. Though memory efficient, convolutional capsules impose geometric constraints which fundamentally limit the ability of capsules to model the pose/deformation of objects. Further, they do not address the bigger memory concern of class-capsules scaling-up to bigger tasks such as detection or large-scale classification. In this study, we introduce deformable capsules (DeformCaps), a new capsule structure (SplitCaps), and a novel dynamic routing algorithm (SE-Routing) to balance computational efficiency with the need for modeling a large number of objects and classes. We demonstrate that the proposed methods allow capsules to efficiently scale-up to large-scale computer vision tasks for the first time, and create the first-ever capsule network for object detection in the literature. Our proposed architecture is a one-stage detection framework and obtains results on MS COCO which are on-par with state-of-the-art one-stage CNN-based methods, while producing fewer false positive detections. | Reject |
ICLR.cc/2019/Conference | Dimensionality Reduction for Representing the Knowledge of Probabilistic Models | Most deep learning models rely on expressive high-dimensional representations to achieve good performance on tasks such as classification. However, the high dimensionality of these representations makes them difficult to interpret and prone to over-fitting. We propose a simple, intuitive and scalable dimension reduction framework that takes into account the soft probabilistic interpretation of standard deep models for classification. When applying our framework to visualization, our representations more accurately reflect inter-class distances than standard visualization techniques such as t-SNE. We show experimentally that our framework improves generalization performance to unseen categories in zero-shot learning. We also provide a finite sample error upper bound guarantee for the method. | Accept (Poster) |
ICLR.cc/2019/Conference | Pix2Scene: Learning Implicit 3D Representations from Images | Modelling 3D scenes from 2D images is a long-standing problem in computer vision with implications in, e.g., simulation and robotics. We propose pix2scene, a deep generative-based approach that implicitly models the geometric properties of a scene from images. Our method learns the depth and orientation of scene points visible in images. Our model can then predict the structure of a scene from various, previously unseen view points. It relies on a bi-directional adversarial learning mechanism to generate scene representations from a latent code, inferring the 3D representation of the underlying scene geometry. We showcase a novel differentiable renderer to train the 3D model in an end-to-end fashion, using only images. We demonstrate the generative ability of our model qualitatively on both a custom dataset and on ShapeNet. Finally, we evaluate the effectiveness of the learned 3D scene representation in supporting a 3D spatial reasoning. | Reject |
ICLR.cc/2023/Conference | Differentially Private $L_2$-Heavy Hitters in the Sliding Window Model | The data management of large companies often prioritize more recent data, as a source of higher accuracy prediction than outdated data. For example, the Facebook data policy retains user search histories for $6$ months while the Google data retention policy states that browser information may be stored for up to $9$ months. These policies are captured by the sliding window model, in which only the most recent $W$ statistics form the underlying dataset. In this paper, we consider the problem of privately releasing the $L_2$-heavy hitters in the sliding window model, which include $L_p$-heavy hitters for $p\le 2$ and in some sense are the strongest possible guarantees that can be achieved using polylogarithmic space, but cannot be handled by existing techniques due to the sub-additivity of the $L_2$ norm. Moreover, existing non-private sliding window algorithms use the smooth histogram framework, which has high sensitivity. To overcome these barriers, we introduce the first differentially private algorithm for $L_2$-heavy hitters in the sliding window model by initiating a number of $L_2$-heavy hitter algorithms across the stream with significantly lower threshold. Similarly, we augment the algorithms with an approximate frequency tracking algorithm with significantly higher accuracy. We then use smooth sensitivity and statistical distance arguments to show that we can add noise proportional to an estimation of the $L_2$ norm. To the best of our knowledge, our techniques are the first to privately release statistics that are related to a sub-additive function in the sliding window model, and may be of independent interest to future differentially private algorithmic design in the sliding window model. | Accept: notable-top-25% |
ICLR.cc/2022/Conference | Degradation Attacks on Certifiably Robust Neural Networks | Certifiably robust neural networks employ provable run-time defenses against adversarial examples by checking if the model is locally robust at the input under evaluation. We show through examples and experiments that these defenses are inherently over-cautious. Specifically, they flag inputs for which local robustness checks fail, but yet that are not adversarial; i.e., they are classified consistently with all valid inputs within a distance of $\epsilon$. As a result, while a norm-bounded adversary cannot change the classification of an input, it can use norm-bounded changes to degrade the utility of certifiably robust networks by forcing them to reject otherwise correctly classifiable inputs. We empirically demonstrate the efficacy of such attacks against state-of-the-art certifiable defenses. | Reject |
ICLR.cc/2023/Conference | Policy Contrastive Imitation Learning | Adversarial imitation learning (AIL) is a popular method that has recently achieved much success. However, the performance of AIL is still unsatisfactory on the more challenging tasks. We find that one of the major reasons is due to the low quality of AIL discriminator representation. Since the AIL discriminator is trained via binary classification that does not necessarily discriminate the policy from the expert in a meaningful way, the resulting reward might not be meaningful either. We propose a new method called Policy Contrastive Imitation Learning (PCIL) to resolve this issue. PCIL learns a contrastive representation space by anchoring on different policies and uses a smooth cosine-similarity-based reward to encourage imitation learning. Our proposed representation learning objective can be viewed as a stronger version of the AIL objective and provide a more meaningful comparison between the agent and the policy. From a theoretical perspective, we show the validity of our method using the apprenticeship learning framework. Furthermore, our empirical evaluation on the DeepMind Control suite demonstrates that PCIL can achieve state-of-the-art performance. Finally, qualitative results suggest that PCIL builds a smoother and more meaningful representation space for imitation learning. | Reject |
ICLR.cc/2023/Conference | Efficient, Stable, and Analytic Differentiation of the Sinkhorn Loss | Optimal transport and the Wasserstein distance have become indispensable building blocks of modern deep generative models, but their computional costs greatly prohibit their applications in statistical machine learning models. Recently, the Sinkhorn loss, as an approximation to the Wasserstein distance, has gained massive popularity, and much work has been done for its theoretical properties. To embed the Sinkhorn loss into gradient-based learning frameworks, efficient algorithms for both the forward and backward passes of the Sinkhorn loss are required. In this article, we first demonstrate issues of the widely-used Sinkhorn's algorithm, and show that the L-BFGS algorithm is a potentially better candidate for the forward pass. Then we derive an analytic form of the derivative of the Sinkhorn loss with respect to the input cost matrix, which results in a very efficient backward algorithm. We rigorously analyze the convergence and stability properties of the advocated algorithms, and use various numerical experiments to validate the superior performance of the proposed methods. | Reject |
ICLR.cc/2021/Conference | Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models | Massively multilingual models subsuming tens or even hundreds of languages pose great challenges to multi-task optimization. While it is a common practice to apply a language-agnostic procedure optimizing a joint multilingual task objective, how to properly characterize and take advantage of its underlying problem structure for improving optimization efficiency remains under-explored. In this paper, we attempt to peek into the black-box of multilingual optimization through the lens of loss function geometry. We find that gradient similarity measured along the optimization trajectory is an important signal, which correlates well with not only language proximity but also the overall model performance. Such observation helps us to identify a critical limitation of existing gradient-based multi-task learning methods, and thus we derive a simple and scalable optimization procedure, named Gradient Vaccine, which encourages more geometrically aligned parameter updates for close tasks. Empirically, our method obtains significant model performance gains on multilingual machine translation and XTREME benchmark tasks for multilingual language models. Our work reveals the importance of properly measuring and utilizing language proximity in multilingual optimization, and has broader implications for multi-task learning beyond multilingual modeling. | Accept (Spotlight) |
ICLR.cc/2023/Conference | Defending against Reconstruction attacks using Rényi Differential Privacy | Reconstruction attacks allow an adversary to regenerate data samples of the training set using access to only a trained model. It has been recently shown that simple heuristics can reconstruct data samples from language models, making this threat scenario an important aspect of model release. Differential privacy is a known solution to such attacks, but is often used with a large privacy budget (epsilon > 8) which does not translate to meaningful guarantees. In this paper we show that, for a same mechanism, we can derive privacy guarantees for reconstruction attacks that are better than the traditional ones from the literature. In particular, we show that larger privacy budgets do not provably protect against membership inference, but can still protect extraction of rare secrets. We design a method to efficiently run reconstruction attacks with lazy sampling and empirically show that we can surface at-risk training samples from non-private language models. We show experimentally that our guarantees hold on real-life language models trained with differential privacy for difficult scenarios, including GPT-2 finetuned on Wikitext-103. | Reject |
ICLR.cc/2020/Conference | Finding Deep Local Optima Using Network Pruning | Artificial neural networks (ANNs) are very popular nowadays and offer reliable solutions to many classification problems. However, training deep neural networks (DNN) is time-consuming due to the large number of parameters. Recent research indicates that these DNNs might be over-parameterized and different solutions have been proposed to reduce the complexity both in the number of parameters and in the training time of the neural networks. Furthermore, some researchers argue that after reducing the neural network complexity via connection pruning, the remaining weights are irrelevant and retraining the sub-network would obtain a comparable accuracy with the original one.
This may hold true in most vision problems where we always enjoy a large number of training samples and research indicates that most local optima of the convolutional neural networks may be equivalent. However, in non-vision sparse datasets, especially with many irrelevant features where a standard neural network would overfit, this might not be the case and there might be many non-equivalent local optima. This paper presents empirical evidence for these statements and an empirical study of the learnability of neural networks (NNs) on some challenging non-linear real and simulated data with irrelevant variables.
Our simulation experiments indicate that the cross-entropy loss function on XOR-like data has many local optima, and the number of local optima grows exponentially with the number of irrelevant variables.
We also introduce a connection pruning method to improve the capability of NNs to find a deep local minimum even when there are irrelevant variables.
Furthermore, the performance of the discovered sparse sub-network degrades considerably either by retraining from scratch or the corresponding original initialization, due to the existence of many bad optima around.
Finally, we will show that the performance of neural networks for real-world experiments on sparse datasets can be recovered or even improved by discovering a good sub-network architecture via connection pruning. | Reject |
ICLR.cc/2020/Conference | Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNets | Skip connections are an essential component of current state-of-the-art deep neural networks (DNNs) such as ResNet, WideResNet, DenseNet, and ResNeXt. Despite their huge success in building deeper and more powerful DNNs, we identify a surprising \emph{security weakness} of skip connections in this paper. Use of skip connections \textit{allows easier generation of highly transferable adversarial examples}. Specifically, in ResNet-like (with skip connections) neural networks, gradients can backpropagate through either skip connections or residual modules. We find that using more gradients from the skip connections rather than the residual modules according to a decay factor, allows one to craft adversarial examples with high transferability. Our method is termed \emph{Skip Gradient Method} (SGM). We conduct comprehensive transfer attacks against state-of-the-art DNNs including ResNets, DenseNets, Inceptions, Inception-ResNet, Squeeze-and-Excitation Network (SENet) and robustly trained DNNs. We show that employing SGM on the gradient flow can greatly improve the transferability of crafted attacks in almost all cases. Furthermore, SGM can be easily combined with existing black-box attack techniques, and obtain high improvements over state-of-the-art transferability methods. Our findings not only motivate new research into the architectural vulnerability of DNNs, but also open up further challenges for the design of secure DNN architectures. | Accept (Spotlight) |
ICLR.cc/2022/Conference | Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation | Since the introduction of the transformer model by Vaswani et al. (2017), a fundamental question has yet to be answered: how does a model achieve extrapolation at inference time for sequences that are longer than it saw during training? We first show that extrapolation can be enabled by simply changing the position representation method, though we find that current methods do not allow for efficient extrapolation. We therefore introduce a simpler and more efficient position method, Attention with Linear Biases (ALiBi). ALiBi does not add positional embeddings to word embeddings; instead, it biases query-key attention scores with a penalty that is proportional to their distance. We show that this method trains a 1.3 billion parameter model on input sequences of length 1024 that extrapolates to input sequences of length 2048, achieving the same perplexity as a sinusoidal position embedding model trained on inputs of length 2048 but training 11% faster and using 11% less memory. ALiBi's inductive bias towards recency also leads it to outperform multiple strong position methods on the WikiText-103 benchmark. | Accept (Poster) |
ICLR.cc/2020/Conference | On the interaction between supervision and self-play in emergent communication | A promising approach for teaching artificial agents to use natural language involves using human-in-the-loop training. However, recent work suggests that current machine learning methods are too data inefficient to be trained in this way from scratch. In this paper, we investigate the relationship between two categories of learning signals with the ultimate goal of improving sample efficiency: imitating human language data via supervised learning, and maximizing reward in a simulated multi-agent environment via self-play (as done in emergent communication), and introduce the term supervised self-play (S2P) for algorithms using both of these signals. We find that first training agents via supervised learning on human data followed by self-play outperforms the converse, suggesting that it is not beneficial to emerge languages from scratch. We then empirically investigate various S2P schedules that begin with supervised learning in two environments: a Lewis signaling game with symbolic inputs, and an image-based referential game with natural language descriptions. Lastly, we introduce population based approaches to S2P, which further improves the performance over single-agent methods. | Accept (Poster) |
ICLR.cc/2023/Conference | Evaluating natural language processing models with generalization metrics that do not need access to any training or testing data | The search for effective and robust metrics has been the focus of recent theoretical and empirical work on generalization of deep neural networks (NNs). In this paper, we discuss the performance of natural language processing (NLP) models, and we evaluate various existing and novel generalization metrics. Compared to prior studies, we (i) focus on NLP instead of computer vision (CV), (ii) focus on generalization metrics that predict test error instead of the generalization gap, (iii) focus on generalization metrics that do not need the access to data, and (iv) focus on the heavy-tail (HT) phenomenon that has received comparatively less attention in the study of deep neural networks. We extend recent HT-based work which focuses on power law (PL) distributions, and we study exponential (EXP) and exponentially truncated power law (E-TPL) fitting to the empirical spectral densities (ESDs) of weight matrices. Our empirical studies are carried on (i) hundreds of Transformers trained in different settings, in which we systematically vary the amount of data, the model size and the optimization hyperparameters, (ii) a total of 51 pretrained Transformers from eight families of Huggingface NLP models, including BERT, GPT2, ALBERT, etc., and (iii) a total of 28 existing and novel generalization metrics. From our detailed empirical analyses, we show that shape metrics, or the metrics obtained from fitting the shape of the ESDs, perform uniformly better at predicting generalization performance than scale metrics commonly studied in the literature, as measured by the average rank correlations with the generalization performance for all of our experiments. We also show that among the three HT distributions considered in our paper, the E-TPL fitting of ESDs performs the most robustly when the models are trained in experimental settings, while the PL fitting achieves the best performance on well-trained Huggingface models, and that both E-TPL and PL metrics (which are both shape metrics) outperform scale metrics. | Reject |
ICLR.cc/2023/Conference | Safe Reinforcement Learning with Contrastive Risk Prediction | As safety violations can lead to severe consequences in real-world applications, the increasing deployment of Reinforcement Learning (RL) in safety-critical domains such as robotics has propelled the study of safe exploration for reinforcement learning (safe RL). In this work, we propose a risk preventive training method for safe RL, which learns a statistical contrastive classifier to predict the probability of a state-action pair leading to unsafe states. Based on the predicted risk probabilities, we can collect risk preventive trajectories and reshape the reward function with risk penalties to induce safe RL policies. We conduct experiments in robotic simulation environments. The results show the proposed approach has comparable performance with the state-of-the-art model-based methods and outperforms conventional model-free safe RL approaches.
| Reject |
ICLR.cc/2023/Conference | Graph Signal Sampling for Inductive One-Bit Matrix Completion: a Closed-form Solution | Inductive one-bit matrix completion is motivated by modern applications such as recommender systems, where new users would appear at test stage with the ratings consisting of only ones and no zeros. We propose a unified graph signal sampling framework which enjoys the benefits of graph signal analysis and processing. The key idea is to transform each user's ratings on the items to a function (signal) on the vertices of an item-item graph, then learn structural graph properties to recover the function from its values on certain vertices --- the problem of graph signal sampling. We propose a class of regularization functionals that takes into account discrete random label noise in the graph vertex domain, then develop the GS-IMC approach which biases the reconstruction towards functions that vary little between adjacent vertices for noise reduction. Theoretical result shows that accurate reconstructions can be achieved under mild conditions. For the online setting, we develop a Bayesian extension, i.e., BGS-IMC which considers continuous random Gaussian noise in the graph Fourier domain and builds upon a prediction-correction update algorithm to obtain the unbiased and minimum-variance reconstruction. Both GS-IMC and BGS-IMC have closed-form solutions and thus are highly scalable in large data. Experiments show that our methods achieve state-of-the-art performance on public benchmarks. | Accept: poster |
ICLR.cc/2019/Conference | Soft Q-Learning with Mutual-Information Regularization | We propose a reinforcement learning (RL) algorithm that uses mutual-information regularization to optimize a prior action distribution for better performance and exploration. Entropy-based regularization has previously been shown to improve both exploration and robustness in challenging sequential decision-making tasks. It does so by encouraging policies to put probability mass on all actions. However, entropy regularization might be undesirable when actions have significantly different importance. In this paper, we propose a theoretically motivated framework that dynamically weights the importance of actions by using the mutual-information. In particular, we express the RL problem as an inference problem where the prior probability distribution over actions is subject to optimization. We show that the prior optimization introduces a mutual-information regularizer in the RL objective. This regularizer encourages the policy to be close to a non-uniform distribution that assigns higher probability mass to more important actions. We empirically demonstrate that our method significantly improves over entropy regularization methods and unregularized methods. | Accept (Poster) |
ICLR.cc/2021/Conference | Attentional Constellation Nets for Few-Shot Learning | The success of deep convolutional neural networks builds on top of the learning of effective convolution operations, capturing a hierarchy of structured features via filtering, activation, and pooling. However, the explicit structured features, e.g. object parts, are not expressive in the existing CNN frameworks. In this paper, we tackle the few-shot learning problem and make an effort to enhance structured features by expanding CNNs with a constellation model, which performs cell feature clustering and encoding with a dense part representation; the relationships among the cell features are further modeled by an attention mechanism. With the additional constellation branch to increase the awareness of object parts, our method is able to attain the advantages of the CNNs while making the overall internal representations more robust in the few-shot learning setting. Our approach attains a significant improvement over the existing methods in few-shot learning on the CIFAR-FS, FC100, and mini-ImageNet benchmarks. | Accept (Poster) |
ICLR.cc/2020/Conference | Plug and Play Language Models: A Simple Approach to Controlled Text Generation | Large transformer-based language models (LMs) trained on huge text corpora have shown unparalleled generation capabilities. However, controlling attributes of the generated language (e.g. switching topic or sentiment) is difficult without modifying the model architecture or fine-tuning on attribute-specific data and entailing the significant cost of retraining. We propose a simple alternative: the Plug and Play Language Model (PPLM) for controllable language generation, which combines a pretrained LM with one or more simple attribute classifiers that guide text generation without any further training of the LM. In the canonical scenario we present, the attribute models are simple classifiers consisting of a user-specified bag of words or a single learned layer with 100,000 times fewer parameters than the LM. Sampling entails a forward and backward pass in which gradients from the attribute model push the LM's hidden activations and thus guide the generation. Model samples demonstrate control over a range of topics and sentiment styles, and extensive automated and human annotated evaluations show attribute alignment and fluency. PPLMs are flexible in that any combination of differentiable attribute models may be used to steer text generation, which will allow for diverse and creative applications beyond the examples given in this paper. | Accept (Poster) |
ICLR.cc/2023/Conference | Laser: Latent Set Representations for 3D Generative Modeling | NeRF provides unparalleled fidelity of novel view synthesis---rendering a 3D scene from an arbitrary viewpoint. NeRF requires training on a large number of views that fully cover a scene, which limits its applicability.
While these issues can be addressed by learning a prior over scenes in various forms, previous approaches have been either applied to overly simple scenes or struggling to render unobserved parts.
We introduce Laser-NV---a generative model which achieves high modelling capacity, and which is based on a set-valued latent representation modelled by normalizing flows.
Similarly to previous amortized approaches, Laser-NV learns structure from multiple scenes and is capable of fast, feed-forward inference from few views.
To encourage higher rendering fidelity and consistency with observed views, Laser-NV further incorporates a geometry-informed attention mechanism over the observed views.
Laser-NV further produces diverse and plausible completions of occluded parts of a scene while remaining consistent with observations.
Laser-NV shows state-of-the-art novel-view synthesis quality when evaluated on ShapeNet and on a novel simulated City dataset, which features high uncertainty in the unobserved regions of the scene. | Reject |
ICLR.cc/2020/Conference | Fast Linear Interpolation for Piecewise-Linear Functions, GAMs, and Deep Lattice Networks | We present fast implementations of linear interpolation operators for both piecewise linear functions and multi-dimensional look-up tables. We use a compiler-based solution (using MLIR) for accelerating this family of workloads. On real-world multi-layer lattice models and a standard CPU, we show these strategies deliver $5-10\times$ faster runtimes compared to a C++ interpreter implementation that uses prior techniques, producing runtimes that are 1000s of times faster than TensorFlow 2.0 for single evaluations. | Reject |
ICLR.cc/2023/Conference | Conformal Prediction is Robust to Label Noise | We study the robustness of conformal prediction—a powerful tool for uncertainty quantification—to label noise. Our analysis tackles both regression and classification problems, characterizing when and how it is possible to construct uncertainty sets that correctly cover the unobserved noiseless ground truth labels. Through stylized theoretical examples and practical experiments, we argue that naïve conformal prediction covers the noiseless ground truth label unless the noise distribution is adversarially designed. This leads us to believe that correcting for label noise is unnecessary except for pathological data distributions or noise sources. In such cases, we can also correct for noise of bounded size in the conformal prediction algorithm in order to ensure correct coverage of the ground truth labels without score or data regularity. | Reject |
ICLR.cc/2023/Conference | Distilling Cognitive Backdoor Patterns within an Image | This paper proposes a simple method to distill and detect backdoor patterns within an image: \emph{Cognitive Distillation} (CD). The idea is to extract the ``minimal essence" from an input image responsible for the model's prediction. CD optimizes an input mask to extract a small pattern from the input image that can lead to the same model output (i.e., logits or deep features). The extracted pattern can help understand the cognitive mechanism of a model on clean vs. backdoor images and is thus called a \emph{Cognitive Pattern} (CP). Using CD and the distilled CPs, we uncover an interesting phenomenon of backdoor attacks: despite the various forms and sizes of trigger patterns used by different attacks, the CPs of backdoor samples are all surprisingly and suspiciously small.
One thus can leverage the learned mask to detect and remove backdoor examples from poisoned training datasets.
We conduct extensive experiments to show that CD can robustly detect a wide range of advanced backdoor attacks.
We also show that CD can potentially be applied to help detect potential biases from face datasets.
Code is available at https://github.com/HanxunH/CognitiveDistillation. | Accept: poster |
ICLR.cc/2023/Conference | SQA3D: Situated Question Answering in 3D Scenes | We propose a new task to benchmark scene understanding of embodied agents: Situated Question Answering in 3D Scenes (SQA3D). Given a scene context (e.g., 3D scan), SQA3D requires the tested agent to first understand its situation (position, orientation, etc.) in the 3D scene as described by text, then reason about its surrounding environment and answer a question under that situation. Based upon 650 scenes from ScanNet, we provide a dataset centered around 6.8k unique situations, along with 20.4k descriptions and 33.4k diverse reasoning questions for these situations. These questions examine a wide spectrum of reasoning capabilities for an intelligent agent, ranging from spatial relation comprehension to commonsense understanding, navigation, and multi-hop reasoning. SQA3D imposes a significant challenge to current multi-modal especially 3D reasoning models. We evaluate various state-of-the-art approaches and find that the best one only achieves an overall score of 47.20%, while amateur human participants can reach 90.06%. We believe SQA3D could facilitate future embodied AI research with stronger situation understanding and reasoning capability. | Accept: poster |
ICLR.cc/2021/Conference | Anchor & Transform: Learning Sparse Embeddings for Large Vocabularies | Learning continuous representations of discrete objects such as text, users, movies, and URLs lies at the heart of many applications including language and user modeling. When using discrete objects as input to neural networks, we often ignore the underlying structures (e.g., natural groupings and similarities) and embed the objects independently into individual vectors. As a result, existing methods do not scale to large vocabulary sizes. In this paper, we design a simple and efficient embedding algorithm that learns a small set of anchor embeddings and a sparse transformation matrix. We call our method Anchor & Transform (ANT) as the embeddings of discrete objects are a sparse linear combination of the anchors, weighted according to the transformation matrix. ANT is scalable, flexible, and end-to-end trainable. We further provide a statistical interpretation of our algorithm as a Bayesian nonparametric prior for embeddings that encourages sparsity and leverages natural groupings among objects. By deriving an approximate inference algorithm based on Small Variance Asymptotics, we obtain a natural extension that automatically learns the optimal number of anchors instead of having to tune it as a hyperparameter. On text classification, language modeling, and movie recommendation benchmarks, we show that ANT is particularly suitable for large vocabulary sizes and demonstrates stronger performance with fewer parameters (up to 40x compression) as compared to existing compression baselines. | Accept (Poster) |
ICLR.cc/2022/Conference | A Zest of LIME: Towards Architecture-Independent Model Distances | Definitions of the distance between two machine learning models either characterize the similarity of the models' predictions or of their weights. While similarity of weights is attractive because it implies similarity of predictions in the limit, it suffers from being inapplicable to comparing models with different architectures. On the other hand, the similarity of predictions is broadly applicable but depends heavily on the choice of model inputs during comparison. In this paper, we instead propose to compute distance between black-box models by comparing their Local Interpretable Model-Agnostic Explanations (LIME). To compare two models, we take a reference dataset, and locally approximate the models on each reference point with linear models trained by LIME. We then compute the cosine distance between the concatenated weights of the linear models. This yields an approach that is both architecture-independent and possesses the benefits of comparing models in weight space. We empirically show that our method, which we call Zest, can be applied to two problems that require measurements of model similarity: detecting model stealing and machine unlearning. | Accept (Poster) |
ICLR.cc/2019/Conference | Episodic Curiosity through Reachability | Rewards are sparse in the real world and most of today's reinforcement learning algorithms struggle with such sparsity. One solution to this problem is to allow the agent to create rewards for itself - thus making rewards dense and more suitable for learning. In particular, inspired by curious behaviour in animals, observing something novel could be rewarded with a bonus. Such bonus is summed up with the real task reward - making it possible for RL algorithms to learn from the combined reward. We propose a new curiosity method which uses episodic memory to form the novelty bonus. To determine the bonus, the current observation is compared with the observations in memory. Crucially, the comparison is done based on how many environment steps it takes to reach the current observation from those in memory - which incorporates rich information about environment dynamics. This allows us to overcome the known "couch-potato" issues of prior work - when the agent finds a way to instantly gratify itself by exploiting actions which lead to hardly predictable consequences. We test our approach in visually rich 3D environments in ViZDoom, DMLab and MuJoCo. In navigational tasks from ViZDoom and DMLab, our agent outperforms the state-of-the-art curiosity method ICM. In MuJoCo, an ant equipped with our curiosity module learns locomotion out of the first-person-view curiosity only. The code is available at https://github.com/google-research/episodic-curiosity/. | Accept (Poster) |
ICLR.cc/2022/Conference | Generalization of Neural Combinatorial Solvers Through the Lens of Adversarial Robustness | End-to-end (geometric) deep learning has seen first successes in approximating the solution of combinatorial optimization problems. However, generating data in the realm of NP-hard/-complete tasks brings practical and theoretical challenges, resulting in evaluation protocols that are too optimistic. Specifically, most datasets only capture a simpler subproblem and likely suffer from spurious features. We investigate these effects by studying adversarial robustness -a local generalization property- to reveal hard, model-specific instances and spurious features. For this purpose, we derive perturbation models for SAT and TSP. Unlike in other applications, where perturbation models are designed around subjective notions of imperceptibility, our perturbation models are efficient and sound, allowing us to determine the true label of perturbed samples without a solver. Surprisingly, with such perturbations, a sufficiently expressive neural solver does not suffer from the limitations of the accuracy-robustness trade-off common in supervised learning. Although such robust solvers exist, we show empirically that the assessed neural solvers do not generalize well w.r.t. small perturbations of the problem instance. | Accept (Poster) |
ICLR.cc/2019/Conference | Trellis Networks for Sequence Modeling | We present trellis networks, a new architecture for sequence modeling. On the one hand, a trellis network is a temporal convolutional network with special structure, characterized by weight tying across depth and direct injection of the input into deep layers. On the other hand, we show that truncated recurrent networks are equivalent to trellis networks with special sparsity structure in their weight matrices. Thus trellis networks with general weight matrices generalize truncated recurrent networks. We leverage these connections to design high-performing trellis networks that absorb structural and algorithmic elements from both recurrent and convolutional models. Experiments demonstrate that trellis networks outperform the current state of the art methods on a variety of challenging benchmarks, including word-level language modeling and character-level language modeling tasks, and stress tests designed to evaluate long-term memory retention. The code is available at https://github.com/locuslab/trellisnet . | Accept (Poster) |
ICLR.cc/2023/Conference | Clifford Neural Layers for PDE Modeling | Partial differential equations (PDEs) see widespread use in sciences and engineering to describe simulation of physical processes as scalar and vector fields interacting and coevolving over time. Due to the computationally expensive nature of their standard solution methods, neural PDE surrogates have become an active research topic to accelerate these simulations. However, current methods do not explicitly take into account the relationship between different fields and their internal components, which are often correlated. Viewing the time evolution of such correlated fields through the lens of multivector fields allows us to overcome these limitations. Multivector fields consist of scalar, vector, as well as higher-order components, such as bivectors and trivectors. Their algebraic properties, such as multiplication, addition and other arithmetic operations can be described by Clifford algebras. To our knowledge, this paper presents the first usage of such multivector representations together with Clifford convolutions and Clifford Fourier transforms in the context of deep learning. The resulting Clifford neural layers are universally applicable and will find direct use in the areas of fluid dynamics, weather forecasting, and the modeling of physical systems in general. We empirically evaluate the benefit of Clifford neural layers by replacing convolution and Fourier operations in common neural PDE surrogates by their Clifford counterparts on 2D Navier-Stokes and weather modeling tasks, as well as 3D Maxwell equations. For similar parameter count, Clifford neural layers consistently improve generalization capabilities of the tested neural PDE surrogates. | Accept: poster |
ICLR.cc/2021/Conference | Towards Defending Multiple Adversarial Perturbations via Gated Batch Normalization | There is now extensive evidence demonstrating that deep neural networks are vulnerable to adversarial examples, motivating the development of defenses against adversarial attacks. However, existing adversarial defenses typically improve model robustness against individual specific perturbation types. Some recent methods improve model robustness against adversarial attacks in multiple $\ell_p$ balls, but their performance against each perturbation type is still far from satisfactory. To better understand this phenomenon, we propose the \emph{multi-domain} hypothesis, stating that different types of adversarial perturbations are drawn from different domains. Guided by the multi-domain hypothesis, we propose~\emph{Gated Batch Normalization (GBN)}, a novel building block for deep neural networks that improves robustness against multiple perturbation types. GBN consists of a gated sub-network and a multi-branch batch normalization (BN) layer, where the gated sub-network separates different perturbation types, and each BN branch is in charge of a single perturbation type and learns domain-specific statistics for input transformation. Then, features from different branches are aligned as domain-invariant representations for the subsequent layers. We perform extensive evaluations of our approach on MNIST, CIFAR-10, and Tiny-ImageNet, and in doing so demonstrate that GBN outperforms previous defense proposals against multiple perturbation types, \ie, $\ell_1$, $\ell_2$, and $\ell_{\infty}$ perturbations, by large margins of 10-20\%. | Reject |
ICLR.cc/2021/Conference | Autonomous Learning of Object-Centric Abstractions for High-Level Planning | We propose a method for autonomously learning an object-centric representation of a continuous and high-dimensional environment that is suitable for planning. Such representations can immediately be transferred between tasks that share the same types of objects, resulting in agents that require fewer samples to learn a model of a new task. We first demonstrate our approach on a simple domain where the agent learns a compact, lifted representation that generalises across objects. We then apply it to a series of Minecraft tasks to learn object-centric representations, including object types—directly from pixel data—that can be leveraged to solve new tasks quickly. The resulting learned representations enable the use of a task-level planner, resulting in an agent capable of forming complex, long-term plans with considerably fewer environment interactions. | Reject |
ICLR.cc/2020/Conference | Weakly Supervised Disentanglement with Guarantees | Learning disentangled representations that correspond to factors of variation in real-world data is critical to interpretable and human-controllable machine learning. Recently, concerns about the viability of learning disentangled representations in a purely unsupervised manner has spurred a shift toward the incorporation of weak supervision. However, there is currently no formalism that identifies when and how weak supervision will guarantee disentanglement. To address this issue, we provide a theoretical framework to assist in analyzing the disentanglement guarantees (or lack thereof) conferred by weak supervision when coupled with learning algorithms based on distribution matching. We empirically verify the guarantees and limitations of several weak supervision methods (restricted labeling, match-pairing, and rank-pairing), demonstrating the predictive power and usefulness of our theoretical framework. | Accept (Poster) |
ICLR.cc/2023/Conference | Linear Video Transformer with Feature Fixation | Vision Transformers have achieved impressive performance in video classification, while suffering from the quadratic complexity caused by the Softmax attention mechanism. Some studies alleviate the computational costs by reducing the number of tokens attended in attention calculation, but the complexity is still quadratic. Another promising way is to replace Softmax attention with linear attention, which owns linear complexity but presents a clear performance drop. We find that such a drop in linear attention results from the lack of attention concentration to critical features. Therefore, we propose a feature fixation module to reweight feature importance of the query and key prior to computing linear attention. Specifically, we regard the query, key, and value as latent representations of the input token, and learn the feature fixation ratio by aggregating Query-Key-Value information. This is beneficial for measuring the feature importance comprehensively. Furthermore, we improve the feature fixation by neighborhood association, which leverages additional guidance from spatial and temporal neighboring tokens. Our proposed method significantly improves the linear attention baseline, and achieves state-of-the-art performance among linear video Transformers on three popular video classification benchmarks. Our performance is even comparable to some quadratic Transformers with fewer parameters and higher efficiency. | Reject |
ICLR.cc/2019/Conference | Evaluating Robustness of Neural Networks with Mixed Integer Programming | Neural networks trained only to optimize for training accuracy can often be fooled by adversarial examples --- slightly perturbed inputs misclassified with high confidence. Verification of networks enables us to gauge their vulnerability to such adversarial examples. We formulate verification of piecewise-linear neural networks as a mixed integer program. On a representative task of finding minimum adversarial distortions, our verifier is two to three orders of magnitude quicker than the state-of-the-art. We achieve this computational speedup via tight formulations for non-linearities, as well as a novel presolve algorithm that makes full use of all information available. The computational speedup allows us to verify properties on convolutional and residual networks with over 100,000 ReLUs --- several orders of magnitude more than networks previously verified by any complete verifier. In particular, we determine for the first time the exact adversarial accuracy of an MNIST classifier to perturbations with bounded l-∞ norm ε=0.1: for this classifier, we find an adversarial example for 4.38% of samples, and a certificate of robustness to norm-bounded perturbations for the remainder. Across all robust training procedures and network architectures considered, and for both the MNIST and CIFAR-10 datasets, we are able to certify more samples than the state-of-the-art and find more adversarial examples than a strong first-order attack. | Accept (Poster) |
ICLR.cc/2020/Conference | Support-guided Adversarial Imitation Learning | We propose Support-guided Adversarial Imitation Learning (SAIL), a generic imitation learning framework that unifies support estimation of the expert policy with the family of Adversarial Imitation Learning (AIL) algorithms. SAIL addresses two important challenges of AIL, including the implicit reward bias and potential training instability. We also show that SAIL is at least as efficient as standard AIL. In an extensive evaluation, we demonstrate that the proposed method effectively handles the reward bias and achieves better performance and training stability than other baseline methods on a wide range of benchmark control tasks. | Reject |
ICLR.cc/2021/Conference | Distribution-Based Invariant Deep Networks for Learning Meta-Features | Recent advances in deep learning from probability distributions successfully achieve classification or regression from distribution samples, thus invariant under permutation of the samples. The first contribution of the paper is to extend these neural architectures to achieve invariance under permutation of the features, too. The proposed architecture, called Dida, inherits the NN properties of universal approximation, and its robustness with respect to Lipschitz-bounded transformations of the input distribution is established. The second contribution is to empirically and comparatively demonstrate the merits of the approach on two tasks defined at the dataset level. On both tasks, Dida learns meta-features supporting the characterization of a (labelled) dataset. The first task consists of predicting whether two dataset patches are extracted from the same initial dataset. The second task consists of predicting whether the learning performance achieved by a hyper-parameter configuration under a fixed algorithm (ranging in k-NN, SVM, logistic regression and linear SGD) dominates that of another configuration, for a dataset extracted from the OpenML benchmarking suite. On both tasks, Dida outperforms the state of the art: DSS and Dataset2Vec architectures, as well as the models based on the hand-crafted meta-features of the literature. | Reject |
ICLR.cc/2022/Conference | Logit Attenuating Weight Normalization | Over-parameterized deep networks trained using gradient-based optimizers is a popular way of solving classification and ranking problems. Without appropriately tuned regularization, such networks have the tendency to make output scores (logits) and network weights large, causing training loss to become too small and the network to lose its adaptivity (ability to move around and escape regions of poor generalization) in the weight space. Adaptive optimizers like Adam, being aggressive at optimizing the train loss, are particularly affected by this. It is well known that, even with weight decay (WD) and normal hyper-parameter tuning, adaptive optimizers lag behind SGD a lot in terms of generalization performance, mainly in the image classification domain.
An alternative to WD for improving a network's adaptivity is to directly control the magnitude of the weights and hence the logits. We propose a method called Logit Attenuating Weight Normalization (LAWN), that can be stacked onto any gradient-based optimizer. LAWN initially starts off training in a free (unregularized) mode and, after some initial epochs, it constrains the weight norms of layers, thereby controlling the logits and improving adaptivity. This is a new regularization approach that does not use WD anywhere; instead, the number of initial free epochs becomes the new hyper-parameter. The resulting LAWN variant of adaptive optimizers gives a solid lift to generalization performance, making their performance equal or even exceed SGD's performance on benchmark image classification and recommender datasets. Another important feature is that LAWN also greatly improves the adaptive optimizers when used with large batch sizes.
| Reject |
ICLR.cc/2023/Conference | Block-level Stiffness Analysis of Residual Networks | Residual Networks (ResNets) can be interpreted as dynamic systems, which are systems whose state changes over time and can be described with ordinary differential equations (ODEs). Specifically, the dynamic systems interpretation views individual residual blocks as ODEs. The solution to an ODE is an approximation; and therefore contains an error term. If an ODE is stiff it is likely that this error is amplified and becomes dominating in the solution calculations, which negatively affects the accuracy of the approximated solution. Therefore, stiff ODEs are often numerically unstable. In this paper we leverage the dynamic systems interpretation to perform a novel theoretical analysis of ResNets by leveraging findings and tools from numerical analysis of ODEs. Specifically, we perform block level stiffness analysis of ResNets. We find that residual blocks towards the end of ResNet models exhibit increased stiffness and that there is a significant correlation between stiffness and model accuracy and loss. Based on these findings, we propose that ResNets behave as stiff numerically unstable ODEs. | Reject |
ICLR.cc/2022/Conference | On the regularization landscape for the linear recommendation models | Recently, a wide range of recommendation algorithms inspired by deep learning techniques have emerged as the performance leaders several standard recommendation benchmarks. While these algorithms were built on different DL techniques (e.g., dropouts, autoencoder), they have similar performance and even similar cost functions. This paper studies whether the models' comparable performance are sheer coincidence, or they can be unified into a single framework. We find that all linear performance leaders effectively add only a nuclear-norm based regularizer, or a Frobenius-norm based regularizer. The former ones possess a (surprisnig) rigid structure that limits the models' predictive power but their solutions are low rank and have closed form. The latter ones are more expressive and more efficient for recommendation but their solutions are either full-rank or require executing hard-to-tune numeric procedures such as ADMM. Along this line of finding, we further propose two low-rank, closed-form solutions, derived from carefully generalizing Frobenius-norm based regularizers. The new solutions get the best of both nuclear-norm and Frobenius-norm world. | Reject |
ICLR.cc/2023/Conference | FAIRER: Fairness as Decision Rationale Alignment | Deep neural networks (DNNs) have achieved remarkable accuracy, but they often suffer from fairness issues, as deep models typically show distinct accuracy differences among some specific subgroups (e.g., males and females). Existing research addresses this critical issue by employing fairness-aware loss functions to constrain the last-layer outputs and directly regularize DNNs. Although the fairness of DNNs is improved, it is unclear how the trained network makes a fair prediction, which limits future fairness improvements. In this paper, we investigate fairness from the perspective of decision rationale and define neuron parity scores to characterize the fair decision process of networks by analyzing neuron behaviors in various subgroups. Extensive empirical studies show that the unfair issue could arise from the unaligned decision rationales of subgroups. Existing fairness regularization terms fail to achieve decision rationale alignment because they only constrain last-layer outputs while ignoring intermediate neuron alignment. To address the issue, we formulate the fairness as a new task, i.e., decision rationale alignment that requires DNNs’ neurons to have consistent responses on subgroups at both intermediate processes and the final prediction. To make this idea practical during optimization, we relax the naive objective function and propose gradient-guided parity alignment, which encourages gradient-weighted consistency of neurons across subgroups. Extensive experiments on a variety of datasets show that our method can improve fairness while maintaining high accuracy and outperforming other baselines by a large margin. We have released our codes at https://anonymous.4open.science/r/fairer_submission-F176/. | Reject |
ICLR.cc/2023/Conference | Towards Estimating Transferability using Hard Subsets | As transfer learning techniques are increasingly used to transfer knowledge from the source model to the target task, it becomes important to quantify which source models are suitable for a given target task without performing computationally expensive fine-tuning. In this work, we propose HASTE (HArd Subset TransfErability), a new strategy to estimate the transferability of a source model to a particular target task using only a harder subset of target data. By leveraging the model’s internal and output representations, we introduce two techniques – one class-agnostic and another class-specific – to identify harder subsets and show that HASTE can be used with any existing transferability metric to improve their reliability. We further analyze the relation between HASTE and the optimal average log-likelihood as well as negative conditional entropy and empirically validate our theoretical bounds. Our experimental results across multiple source model architectures, target datasets, and transfer learning tasks show that HASTE-modified metrics are consistently better or on par with the state-of-the-art transferability metrics. | Reject |
ICLR.cc/2022/Conference | Tesseract: Gradient Flip Score to Secure Federated Learning against Model Poisoning Attacks | Federated learning—multi-party, distributed learning in a decentralized environment—is vulnerable to model poisoning attacks, even more so than centralized learning approaches. This is because malicious clients can collude and send in carefully tailored model updates to make the global model inaccurate. This motivated the development of Byzantine-resilient federated learning algorithms, such as Krum, Trimmed mean, and FoolsGold. However, a recently developed targeted model poisoning attack showed that all prior defenses can be bypassed. The attack uses the intuition that simply by changing the sign of the gradient updates that the optimizer is computing, for a set of malicious clients, a model can be pushed away from the optima to increase the test error rate. In this work, we develop tesseract—a defense against this directed deviation attack, a state-of-the-art model poisoning attack. TESSERACT is based on a simple intuition that in a federated learning setting, certain patterns of gradient flips are indicative of an attack. This intuition is remarkably stable across different learning algorithms, models, and datasets. TESSERACT assigns reputation scores to the participating clients based on their behavior during the training phase and then takes a weighted contribution of the clients. We show that TESSERACT provides robustness against even an adaptive white-box version of the attack. | Reject |
ICLR.cc/2020/Conference | Making Efficient Use of Demonstrations to Solve Hard Exploration Problems | This paper introduces R2D3, an agent that makes efficient use of demonstrations to solve hard exploration problems in partially observable environments with highly variable initial conditions. We also introduce a suite of eight tasks that combine these three properties, and show that R2D3 can solve several of the tasks where other state of the art methods (both with and without demonstrations) fail to see even a single successful trajectory after tens of billions of steps of exploration. | Accept (Poster) |
ICLR.cc/2022/Conference | Increasing the Cost of Model Extraction with Calibrated Proof of Work | In model extraction attacks, adversaries can steal a machine learning model exposed via a public API by repeatedly querying it and adjusting their own model based on obtained predictions. To prevent model stealing, existing defenses focus on detecting malicious queries, truncating, or distorting outputs, thus necessarily introducing a tradeoff between robustness and model utility for legitimate users. Instead, we propose to impede model extraction by requiring users to complete a proof-of-work before they can read the model's predictions. This deters attackers by greatly increasing (even up to 100x) the computational effort needed to leverage query access for model extraction. Since we calibrate the effort required to complete the proof-of-work to each query, this only introduces a slight overhead for regular users (up to 2x). To achieve this, our calibration applies tools from differential privacy to measure the information revealed by a query. Our method requires no modification of the victim model and can be applied by machine learning practitioners to guard their publicly exposed models against being easily stolen. | Accept (Spotlight) |
ICLR.cc/2023/Conference | Generating Diverse Cooperative Agents by Learning Incompatible Policies | Training a robust cooperative agent requires diverse partner agents. However, obtaining those agents is difficult. Previous works aim to learn diverse behaviors by changing the state-action distribution of agents. But, without information about the task's goal, the diversified agents are not guided to find other important, albeit sub-optimal, solutions: the agents might learn only variations of the same solution. In this work, we propose to learn diverse behaviors via policy compatibility. Conceptually, policy compatibility measures whether policies of interest can coordinate effectively. We theoretically show that incompatible policies are not similar. Thus, policy compatibility—which has been used exclusively as a measure of robustness—can be used as a proxy for learning diverse behaviors. Then, we incorporate the proposed objective into a population-based training scheme to allow concurrent training of multiple agents. Additionally, we use state-action information to induce local variations of each policy. Empirically, the proposed method consistently discovers more solutions than baseline methods across various multi-goal cooperative environments. Finally, in multi-recipe Overcooked, we show that our method produces populations of behaviorally diverse agents, which enables generalist agents trained with such a population to be more robust.
See our project page at https://bit.ly/marl-lipo
| Accept: notable-top-25% |
ICLR.cc/2022/Conference | MaiT: integrating spatial locality into image transformers with attention masks | Though image transformers have shown competitive results with convolutional neural networks in computer vision tasks, lacking inductive biases such as locality still poses problems in terms of model efficiency especially for embedded applications. In this work, we address this issue by introducing attention masks to incorporate spatial locality into self-attention heads of transformers. Local dependencies are captured with masked attention heads along with global dependencies captured by original unmasked attention heads. With Masked attention image Transformer – MaiT, top-1 accuracy increases by up to 1.0\% compared to DeiT, without extra parameters, computation, or external training data. Moreover, attention masks regulate the training of attention maps, which facilitates the convergence and improves the accuracy of deeper transformers. Masked attention heads guide the model to focus on local information in early layers and promote diverse attention maps in latter layers. Deep MaiT improves the top-1 accuracy by up to 1.5\% compared to CaiT with fewer parameters and less FLOPs. Encoding locality with attention masks requires no extra parameter or structural change, and thus it can be combined with other techniques for further improvement in vision transformers. | Reject |
ICLR.cc/2023/Conference | Federated Learning of Large Models at the Edge via Principal Sub-Model Training | Limited compute, memory, and communication capabilities of edge users create a significant bottleneck for federated learning (FL) of large models. Current literature typically tackles the challenge with a heterogeneous client setting or allows training to be offloaded to the server. However, the former requires a fraction of clients to train near-full models, which may not be achievable at the edge; while the latter can compromise privacy with sharing of intermediate representations or labels. In this work, we consider a realistic, but much less explored, cross-device FL setting in which no client has the capacity to train a full large model nor is willing to share any intermediate representations with the server. To this end, we present Principal Sub-Model (PriSM) training methodology, which leverages models’ low-rank structure and kernel orthogonality to train sub-models in the orthogonal kernel space. More specifically, by applying singular value decomposition to original kernels in the server model, PriSM first obtains a set of principal orthogonal kernels with importance weighed by their singular values. Thereafter, PriSM utilizes a novel sampling strategy that selects different subsets of the principal kernels independently to create sub-models for clients with reduced computation and communication requirements. Importantly, a kernel with a large singular value is assigned with a high sampling probability. Thus, each sub-model is a low-rank approximation of the full large model, and all clients together achieve nearly full coverage of the principal kernels. To further improve memory efficiency, PriSM exploits low-rank structure in intermediate representations and allows each sub-model to learn only a subset of them while still preserving training performance. Our extensive evaluations on multiple datasets in various resource-constrained settings demonstrate that PriSM can yield an improved performance of up to $10\%$ compared to existing alternatives, when training sub-models with only $20\%$ principal kernels ($\sim 5\%$ of the full server model.). | Reject |
ICLR.cc/2018/Conference | Fix your classifier: the marginal value of training the last weight layer | Neural networks are commonly used as models for classification for a wide variety of tasks. Typically, a learned affine transformation is placed at the end of such models, yielding a per-class value used for classification. This classifier can have a vast number of parameters, which grows linearly with the number of possible classes, thus requiring increasingly more resources.
In this work we argue that this classifier can be fixed, up to a global scale constant, with little or no loss of accuracy for most tasks, allowing memory and computational benefits. Moreover, we show that by initializing the classifier with a Hadamard matrix we can speed up inference as well. We discuss the implications for current understanding of neural network models.
| Accept (Poster) |
ICLR.cc/2022/Conference | Cross-Domain Imitation Learning via Optimal Transport | Cross-domain imitation learning studies how to leverage expert demonstrations of one agent to train an imitation agent with a different embodiment or morphology. Comparing trajectories and stationary distributions between the expert and imitation agents is challenging because they live on different systems that may not even have the same dimensionality. We propose Gromov-Wasserstein Imitation Learning (GWIL), a method for cross-domain imitation that uses the Gromov-Wasserstein distance to align and compare states between the different spaces of the agents. Our theory formally characterizes the scenarios where GWIL preserves optimality, revealing its possibilities and limitations. We demonstrate the effectiveness of GWIL in non-trivial continuous control domains ranging from simple rigid transformation of the expert domain to arbitrary transformation of the state-action space. | Accept (Poster) |
ICLR.cc/2022/Conference | Improving Long-Horizon Imitation Through Language Prediction | Complex, long-horizon planning and its combinatorial nature pose steep challenges for learning-based agents. Difficulties in such settings are exacerbated in low data regimes where over-fitting stifles generalization and compounding errors hurt accuracy. In this work, we explore the use of an often unused source of auxiliary supervision: language. Inspired by recent advances in transformer-based models, we train agents with an instruction prediction loss that encourages learning temporally extended representations that operate at a high level of abstraction. Concretely, we demonstrate that instruction modeling significantly improves performance in planning environments when training with a limited number of demonstrations on the BabyAI and Crafter benchmarks. In further analysis we find that instruction modeling is most important for tasks that require complex reasoning, while understandably offering smaller gains in environments that require simple plans. Our benchmarks and code will be publicly released. | Reject |
ICLR.cc/2023/Conference | A Graph Neural Network Approach to Automated Model Building in Cryo-EM Maps | Electron cryo-microscopy (cryo-EM) produces three-dimensional (3D) maps of the electrostatic potential of biological macromolecules, including proteins. At sufficient resolution, the cryo-EM maps, along with some knowledge about the imaged molecules, allow de novo atomic modelling. Typically, this is done through a laborious manual process. Recent advances in machine learning applications to protein structure prediction show potential for automating this process. Taking inspiration from these techniques, we have built ModelAngelo for automated model building of proteins in cryo-EM maps. ModelAngelo first uses a residual convolutional neural network (CNN) to initialize a graph representation with nodes assigned to individual amino acids of the proteins in the map and edges representing the protein chain. The graph is then refined with a graph neural network (GNN) that combines the cryo-EM data, the amino acid sequence data and prior knowledge about protein geometries. The GNN refines the geometry of the protein chain and classifies the amino acids for each of its nodes. The final graph is post-processed with a hidden Markov model (HMM) search to map each protein chain to entries in a user provided sequence file. Application to 28 test cases shows that ModelAngelo outperforms state-of-the-art and approximates manual building for cryo-EM maps with resolutions better than 3.5 A. | Accept: poster |
ICLR.cc/2022/Conference | MT-GBM: A Multi-Task Gradient Boosting Machine with Shared Decision Trees | Despite the success of deep learning in computer vision and natural language processing, Gradient Boosted Decision Tree (GBDT) is yet one of the most powerful tools for applications with tabular data such as e-commerce and FinTech. However, applying GBDT to multi-task learning is still a challenge. Unlike deep models that can jointly learn a shared latent representation across multiple tasks, GBDT can hardly learn a shared tree structure.
In this paper, we propose Multi-Task Gradient Boosting Machine (MT-GBM), a GBDT-based method for multi-task learning. The MT-GBM can find the shared tree structures and split branches according to multi-task losses. First, it assigns multiple outputs to each leaf node. Next, it computes the gradient corresponding to each output (task). Then, we also propose an algorithm to combine the gradients of all tasks and update the tree. Finally, we apply MT-GBM to LightGBM. Experiments show that our MT-GBM improves the performance of the main task significantly, which means the proposed MT-GBM is efficient and effective. | Reject |
ICLR.cc/2021/Conference | Model Patching: Closing the Subgroup Performance Gap with Data Augmentation | Classifiers in machine learning are often brittle when deployed. Particularly concerning are models with inconsistent performance on specific subgroups of a class, e.g., exhibiting disparities in skin cancer classification in the presence or absence of a spurious bandage. To mitigate these performance differences, we introduce model patching, a two-stage framework for improving robustness that encourages the model to be invariant to subgroup differences, and focus on class information shared by subgroups. Model patching first models subgroup features within a class and learns semantic transformations between them, and then trains a classifier with data augmentations that deliberately manipulate subgroup features. We instantiate model patching with CAMEL, which (1) uses a CycleGAN to learn the intra-class, inter-subgroup augmentations, and (2) balances subgroup performance using a theoretically-motivated subgroup consistency regularizer, accompanied by a new robust objective. We demonstrate CAMEL’s effectiveness on 3 benchmark datasets, with reductions in robust error of up to 33% relative to the best baseline. Lastly, CAMEL successfully patches a model that fails due to spurious features on a real-world skin cancer dataset. | Accept (Poster) |
ICLR.cc/2022/Conference | Learning to Remember Patterns: Pattern Matching Memory Networks for Traffic Forecasting | Traffic forecasting is a challenging problem due to complex road networks and sudden speed changes caused by various events on roads. Several models have been proposed to solve this challenging problem, with a focus on learning the spatio-temporal dependencies of roads. In this work, we propose a new perspective for converting the forecasting problem into a pattern-matching task, assuming that large traffic data can be represented by a set of patterns. To evaluate the validity of this new perspective, we design a novel traffic forecasting model called Pattern-Matching Memory Networks (PM-MemNet), which learns to match input data to representative patterns with a key-value memory structure. We first extract and cluster representative traffic patterns that serve as keys in the memory. Then, by matching the extracted keys and inputs, PM-MemNet acquires the necessary information on existing traffic patterns from the memory and uses it for forecasting. To model the spatio-temporal correlation of traffic, we proposed a novel memory architecture, GCMem, which integrates attention and graph convolution. The experimental results indicate that PM-MemNet is more accurate than state-of-the-art models, such as Graph WaveNet, with higher responsiveness. We also present a qualitative analysis describing how PM-MemNet works and achieves higher accuracy when road speed changes rapidly. | Accept (Poster) |
ICLR.cc/2021/Conference | Provable Fictitious Play for General Mean-Field Games | We propose a reinforcement learning algorithm for stationary mean-field games, where the goal is to learn a pair of mean-field state and stationary policy that constitutes the Nash equilibrium. When viewing the mean-field state and the policy as two players, we propose a fictitious play algorithm which alternatively updates the mean-field state and the policy via gradient-descent and proximal policy optimization, respectively. Our algorithm is in stark contrast with previous literature which solves each single-agent reinforcement learning problem induced by the iterates mean-field states to the optimum. Furthermore, we prove that our fictitious play algorithm converges to the Nash equilibrium at a sublinear rate. To the best of our knowledge, this seems the first provably convergent reinforcement learning algorithm for mean-field games based on iterative updates of both mean-field state and policy. | Reject |
ICLR.cc/2023/Conference | On Convergence of Average-Reward Off-Policy Control Algorithms in Weakly-Communicating MDPs | We show two average-reward off-policy control algorithms, Differential Q Learning (Wan, Naik, \& Sutton 2021a) and RVI Q Learning (Abounadi Bertsekas \& Borkar 2001), converge in weakly-communicating MDPs. Weakly-communicating MDPs are the most general class of MDPs that a learning algorithm with a single stream of experience can guarantee obtaining a policy achieving optimal reward rate. The original convergence proofs of the two algorithms require that all optimal policies induce unichains, which is not necessarily true for weakly-communicating MDPs. To the best of our knowledge, our results are the first showing average-reward off-policy control algorithms converge in weakly-communicating MDPs. As a direct extension, we show that average-reward options algorithms introduced by (Wan, Naik, \& Sutton 2021b) converge if the Semi-MDP induced by options is weakly-communicating. | Reject |
ICLR.cc/2021/Conference | Ruminating Word Representations with Random Noise Masking | We introduce a training method for better word representation and performance, which we call \textbf{GraVeR} (\textbf{Gra}dual \textbf{Ve}ctor \textbf{R}umination). The method is to gradually and iteratively add random noises and bias to word embeddings after training a model, and re-train the model from scratch but initialize with the noised word embeddings. Through the re-training process, some noises can be compensated and other noises can be utilized to learn better representations. As a result, we can get word representations further fine-tuned and specialized in the task. On six text classification tasks, our method improves model performances with a large gap. When GraVeR is combined with other regularization techniques, it shows further improvements. Lastly, we investigate the usefulness of GraVeR. | Reject |
ICLR.cc/2022/Conference | Information-Aware Time Series Meta-Contrastive Learning | Various contrastive learning approaches have been proposed in recent years and achieve significant empirical success. While effective and prevalent, contrastive learning has been less explored for time series data. A key component of contrastive learning is to select appropriate augmentations imposing some priors to construct feasible positive samples, such that an encoder can be trained to learn robust and discriminative representations. Unlike image and language domains where ``desired'' augmented samples can be generated with the rule of thumb guided by prefabricated human priors, the ad-hoc manual selection of time series augmentations is hindered by their diverse and human-unrecognizable temporal structures. How to find the desired augmentations of time series data that are meaningful for given contrastive learning tasks and datasets remains an open question. In this work, we address the problem by encouraging both high fidelity and variety based upon information theory. A theoretical analysis leads to the criteria for selecting feasible data augmentations. On top of that, we employ the meta-learning mechanism and propose an information-aware approach, InfoTS, that adaptively selects optimal time series augmentations for contrastive representation learning. The meta-learner and the encoder are jointly optimized in an end-to-end manner to avoid sub-optimal solutions. Experiments on various datasets show highly competitive performance with up to 11.4% reduction in MSE on the forecasting task and up to 2.8% relative improvement in accuracy on the classification task over the leading baselines. | Reject |
ICLR.cc/2018/Conference | PrivyNet: A Flexible Framework for Privacy-Preserving Deep Neural Network Training | Massive data exist among user local platforms that usually cannot support deep neural network (DNN) training due to computation and storage resource constraints. Cloud-based training schemes provide beneficial services but suffer from potential privacy risks due to excessive user data collection. To enable cloud-based DNN training while protecting the data privacy simultaneously, we propose to leverage the intermediate representations of the data, which is achieved by splitting the DNNs and deploying them separately onto local platforms and the cloud. The local neural network (NN) is used to generate the feature representations. To avoid local training and protect data privacy, the local NN is derived from pre-trained NNs. The cloud NN is then trained based on the extracted intermediate representations for the target learning task. We validate the idea of DNN splitting by characterizing the dependency of privacy loss and classification accuracy on the local NN topology for a convolutional NN (CNN) based image classification task. Based on the characterization, we further propose PrivyNet to determine the local NN topology, which optimizes the accuracy of the target learning task under the constraints on privacy loss, local computation, and storage. The efficiency and effectiveness of PrivyNet are demonstrated with CIFAR-10 dataset. | Reject |
ICLR.cc/2022/Conference | Learning to Generalize Compositionally by Transferring Across Semantic Parsing Tasks | Neural network models often generalize poorly to mismatched domains or distributions. In NLP, this issue arises in particular when models are expected to generalize compositionally, that is, to novel combinations of familiar words and constructions. We investigate learning representations that facilitate transfer learning from one compositional task to another: the representation and the task-specific layers of the models are strategically trained differently on a pre-finetuning task such that they generalize well on mismatched splits that require compositionality. We apply this method to semantic parsing, using three very different datasets, COGS, GeoQuery and SCAN, used alternately as the pre-finetuning and target task. Our method significantly improves compositional generalization over baselines on the test set of the target task, which is held out during fine-tuning. Ablation studies characterize the utility of the major steps in the proposed algorithm and support our hypothesis. | Reject |
ICLR.cc/2023/Conference | Bag of Tricks for FGSM Adversarial Training | Adversarial training (AT) with samples generated by Fast Gradient Sign Method (FGSM), also known as FGSM-AT, is a computationally simple method to train robust networks. However, during its training procedure, an unstable mode of ``catastrophic overfitting'' has been identified in [Wong2020FastIB], where the robust accuracy abruptly drops to zero within a single training step. Existing methods use gradient regularizers or random initialization tricks to attenuate this issue, whereas they either take high computational cost or lead to lower robust accuracy. In this work, we provide the first study, which thoroughly examines a collection of tricks from three perspectives: Data Initialization, Network Structure, and Optimization, to overcome the catastrophic overfitting in FGSM-AT.
Surprisingly, we find that simple tricks, i.e., a) masking partial pixels (even without randomness), b) setting a large convolution stride and smooth activation functions, or c) regularizing the weights of the first convolutional layer, can effectively tackle the overfitting issue.
Extensive results on a range of network architectures validate the effectiveness of each proposed trick, and the combinations of tricks are also investigated. For example, trained with PreActResNet-18 on CIFAR-10, our method attains 49.8% accuracy against PGD-50 attacker and 46.4% accuracy against AutoAttack, demonstrating that pure FGSM-AT is capable of enabling robust learners. | Reject |
ICLR.cc/2023/Conference | Is Self-Supervised Contrastive Learning More Robust Than Supervised Learning? | Prior work on self-supervised contrastive learning has primarily focused on evaluating the recognition accuracy, but has overlooked other behavioral aspects. In addition to accuracy, distributional robustness plays a critical role in the reliability of machine learning models. We design and conduct a series of robustness tests to quantify the behavioral differences between contrastive learning and supervised learning to downstream and pre-training data distribution changes. These tests leverage data corruptions at multiple levels, ranging from pixel-level distortion to patch-level shuffling and to dataset-level distribution shift, including both natural and unnatural corruptions. Our tests unveil intriguing robustness behaviors of contrastive and supervised learning: while we generally observe that contrastive learning is more robust than supervised learning under downstream corruptions, we surprisingly discover the robustness vulnerability of contrastive learning under pixel and patch level corruptions during pre-training. Furthermore, we observe the higher dependence of contrastive learning on spatial image coherence information during pre-training, e.g., it is particularly sensitive to global patch shuffling. We explain these results by connecting to feature space uniformity and data augmentation. Our analysis has implications in improving the downstream robustness of supervised learning, and calls for more studies on understanding contrastive learning. | Reject |
ICLR.cc/2021/Conference | Early Stopping by Gradient Disparity | Validation-based early-stopping methods are one of the most popular techniques used to avoid over-training deep neural networks. They require to set aside a reliable unbiased validation set, which can be expensive in applications offering limited amounts of data. In this paper, we propose to use \emph{gradient disparity}, which we define as the $\ell_2$ norm distance between the gradient vectors of two batches drawn from the training set. It comes from a probabilistic upper bound on the difference between the classification errors over a given batch, when the network is trained on this batch and when the network is trained on another batch of points sampled from the same dataset. We empirically show that gradient disparity is a very promising early-stopping criterion when data is limited, because it uses all the training samples during training. Furthermore, we show in a wide range of experimental settings that gradient disparity is not only strongly related to the usual generalization error between the training and test sets, but that it is also much more informative about the level of label noise. | Reject |
ICLR.cc/2020/Conference | Multi-source Multi-view Transfer Learning in Neural Topic Modeling with Pretrained Topic and Word Embeddings | Though word embeddings and topics are complementary representations, several
past works have only used pretrained word embeddings in (neural) topic modeling
to address data sparsity problem in short text or small collection of documents.
However, no prior work has employed (pretrained latent) topics in transfer learning
paradigm. In this paper, we propose a framework to perform transfer learning
in neural topic modeling using (1) pretrained (latent) topics obtained from a large
source corpus, and (2) pretrained word and topic embeddings jointly (i.e., multiview)
in order to improve topic quality, better deal with polysemy and data sparsity
issues in a target corpus. In doing so, we first accumulate topics and word representations
from one or many source corpora to build respective pools of pretrained
topic (i.e., TopicPool) and word embeddings (i.e., WordPool). Then, we identify
one or multiple relevant source domain(s) and take advantage of corresponding
topics and word features via the respective pools to guide meaningful learning
in the sparse target domain. We quantify the quality of topic and document representations
via generalization (perplexity), interpretability (topic coherence) and
information retrieval (IR) using short-text, long-text, small and large document
collections from news and medical domains. We have demonstrated the state-ofthe-
art results on topic modeling with the proposed transfer learning approaches. | Reject |
ICLR.cc/2020/Conference | Learning Human Postural Control with Hierarchical Acquisition Functions | Learning control policies in robotic tasks requires a large number of interactions due to small learning rates, bounds on the updates or unknown constraints. In contrast humans can infer protective and safe solutions after a single failure or unexpected observation.
In order to reach similar performance, we developed a hierarchical Bayesian optimization algorithm that replicates the cognitive inference and memorization process for avoiding failures in motor control tasks. A Gaussian Process implements the modeling and the sampling of the acquisition function. This enables rapid learning with large learning rates while a mental replay phase ensures that policy regions that led to failures are inhibited during the sampling process.
The features of the hierarchical Bayesian optimization method are evaluated in a simulated and physiological humanoid postural balancing task. We quantitatively compare the human learning performance to our learning approach by evaluating the deviations of the center of mass during training. Our results show that we can reproduce the efficient learning of human subjects in postural control tasks which provides a testable model for future physiological motor control tasks. In these postural control tasks, our method outperforms standard Bayesian Optimization in the number of interactions to solve the task, in the computational demands and in the frequency of observed failures. | Reject |
ICLR.cc/2020/Conference | Scaling Autoregressive Video Models | Due to the statistical complexity of video, the high degree of inherent stochasticity, and the sheer amount of data, generating natural video remains a challenging task. State-of-the-art video generation models attempt to address these issues by combining sometimes complex, often video-specific neural network architectures, latent variable models, adversarial training and a range of other methods. Despite their often high complexity, these approaches still fall short of generating high quality video continuations outside of narrow domains and often struggle with fidelity. In contrast, we show that conceptually simple, autoregressive video generation models based on a three-dimensional self-attention mechanism achieve highly competitive results across multiple metrics on popular benchmark datasets for which they produce continuations of high fidelity and realism. Furthermore, we find that our models are capable of producing diverse and surprisingly realistic continuations on a subset of videos from Kinetics, a large scale action recognition dataset comprised of YouTube videos exhibiting phenomena such as camera movement, complex object interactions and diverse human movement. To our knowledge, this is the first promising application of video-generation models to videos of this complexity. | Accept (Spotlight) |
ICLR.cc/2018/Conference | On the State of the Art of Evaluation in Neural Language Models | Ongoing innovations in recurrent neural network architectures have provided a steady influx of apparently state-of-the-art results on language modelling benchmarks. However, these have been evaluated using differing codebases and limited computational resources, which represent uncontrolled sources of experimental variation. We reevaluate several popular architectures and regularisation methods with large-scale automatic black-box hyperparameter tuning and arrive at the somewhat surprising conclusion that standard LSTM architectures, when properly regularised, outperform more recent models. We establish a new state of the art on the Penn Treebank and Wikitext-2 corpora, as well as strong baselines on the Hutter Prize dataset.
| Accept (Poster) |
ICLR.cc/2023/Conference | Contextual Symbolic Policy For Meta-Reinforcement Learning | Context-based Meta-Reinforcement Learning (Meta-RL), which conditions the RL agent on the context variables, is a powerful method for learning a generalizable agent.
Current context-based Meta-RL methods often construct their contextual policy with a neural network (NN) and directly take the context variables as a part of the input. However, the NN-based policy contains tremendous parameters which possibly result in overfitting, the difficulty of deployment and poor interpretability.
To improve the generation ability, efficiency and interpretability, we propose a novel Contextual Symbolic Policy (CSP) framework, which generates contextual policy with a symbolic form based on the context variables for unseen tasks in meta-RL. Our key insight is that the symbolic expression is capable of capturing complex relationships by composing various operators and has a compact form that helps strip out irrelevant information. Thus, the CSP learns to produce symbolic policy for meta-RL tasks and extract the essential common knowledge to achieve higher generalization ability. Besides, the symbolic policies with a compact form are efficient to be deployed and easier to understand.
In the implementation, we construct CSP as a gradient-based framework to learn the symbolic policy from scratch in an end-to-end and differentiable way. The symbolic policy is represented by a symbolic network composed of various symbolic operators. We also employ a path selector to decide the proper symbolic form of the policy and a parameter generator to produce the coefficients of the symbolic policy. Empirically, we evaluate the proposed CSP method on several Meta-RL tasks and demonstrate that the contextual symbolic policy achieves higher performance and efficiency and shows the potential to be interpretable. | Reject |
ICLR.cc/2023/Conference | DCT-DiffStride: Differentiable Strides with Real-Valued Data | Reducing the size of intermediate feature maps within various neural network architectures is critical for generalization performance, and memory and computational complexity. Until recently, most methods required downsampling rates (i.e., decimation) to be predefined and static during training, with optimal downsampling rates requiring a vast hyper-parameter search. Recent work has proposed a novel and differentiable method for learning strides named DiffStride which uses the discrete Fourier transform (DFT) to learn strides for decimation. However, in many cases the DFT does not capture signal properties as efficiently as the discrete cosine transform (DCT). Therefore, we propose an alternative method for learning decimation strides, DCT-DiffStride, as well as new regularization methods to reduce model complexity. Our work employs the DCT and its inverse as a low-pass filter in the frequency domain to reduce feature map dimensionality. Leveraging the well-known energy compaction properties of the DCT for natural signals, we evaluate DCT-DiffStride with its competitors on image and audio datasets demonstrating a favorable tradeoff in model performance and model complexity compared to competing methods. Additionally, we show DCT-DiffStride and DiffStride can be applied to data outside the natural signal domain, increasing the general applications of such methods. | Reject |
ICLR.cc/2020/Conference | Sharing Knowledge in Multi-Task Deep Reinforcement Learning | We study the benefit of sharing representations among tasks to enable the effective use of deep neural networks in Multi-Task Reinforcement Learning. We leverage the assumption that learning from different tasks, sharing common properties, is helpful to generalize the knowledge of them resulting in a more effective feature extraction compared to learning a single task. Intuitively, the resulting set of features offers performance benefits when used by Reinforcement Learning algorithms. We prove this by providing theoretical guarantees that highlight the conditions for which is convenient to share representations among tasks, extending the well-known finite-time bounds of Approximate Value-Iteration to the multi-task setting. In addition, we complement our analysis by proposing multi-task extensions of three Reinforcement Learning algorithms that we empirically evaluate on widely used Reinforcement Learning benchmarks showing significant improvements over the single-task counterparts in terms of sample efficiency and performance. | Accept (Poster) |
ICLR.cc/2021/Conference | Deep Repulsive Clustering of Ordered Data Based on Order-Identity Decomposition | We propose the deep repulsive clustering (DRC) algorithm of ordered data for effective order learning. First, we develop the order-identity decomposition (ORID) network to divide the information of an object instance into an order-related feature and an identity feature. Then, we group object instances into clusters according to their identity features using a repulsive term. Moreover, we estimate the rank of a test instance, by comparing it with references within the same cluster. Experimental results on facial age estimation, aesthetic score regression, and historical color image classification show that the proposed algorithm can cluster ordered data effectively and also yield excellent rank estimation performance. | Accept (Poster) |
ICLR.cc/2018/Conference | Revisiting Knowledge Base Embedding as Tensor Decomposition | We study the problem of knowledge base (KB) embedding, which is usually addressed through two frameworks---neural KB embedding and tensor decomposition. In this work, we theoretically analyze the neural embedding framework and subsequently connect it with tensor based embedding. Specifically, we show that in neural KB embedding the two commonly adopted optimization solutions---margin-based and negative sampling losses---are closely related to each other. We also reach the closed-form tensor that is implicitly approximated by popular neural KB approaches, revealing the underlying connection between neural and tensor based KB embedding models. Grounded in the theoretical results, we further present a tensor decomposition based framework KBTD to directly approximate the derived closed form tensor. Under this framework, the neural KB embedding models, such as NTN, TransE, Bilinear, and DISTMULT, are unified into a general tensor optimization architecture. Finally, we conduct experiments on the link prediction task in WordNet and Freebase, empirically demonstrating the effectiveness of the KBTD framework.
| Reject |
ICLR.cc/2022/Conference | TRAKR – A reservoir-based tool for fast and accurate classification of neural time-series patterns | Neuroscience has seen a dramatic increase in the types of recording modalities and complexity of neural time-series data collected from them. The brain is a highly recurrent system producing rich, complex dynamics that result in different behaviors. Correctly distinguishing such nonlinear neural time series in real-time, especially those with non-obvious links to behavior, could be useful for a wide variety of applications. These include detecting anomalous clinical events such as seizures in epilepsy, and identifying optimal control spaces for brain machine interfaces. It remains challenging to correctly distinguish nonlinear time-series patterns because of the high intrinsic dimensionality of such data, making accurate inference of state changes (for intervention or control) difficult. Simple distance metrics, which can be computed quickly do not yield accurate classifications. On the other end of the spectrum of classification methods, ensembles of classifiers or deep supervised tools offer higher accuracy but are slow, data-intensive, and computationally expensive. We introduce a reservoir-based tool, state tracker (TRAKR), which offers the high accuracy of ensembles or deep supervised methods while preserving the computational benefits of simple distance metrics. After one-shot training, TRAKR can accurately, and in real time, detect deviations in test patterns. By forcing the weighted dynamics of the reservoir to fit a desired pattern directly, we avoid many rounds of expensive optimization. Then, keeping the output weights frozen, we use the error signal generated by the reservoir in response to a particular test pattern as a classification boundary. We show that, using this approach, TRAKR accurately detects changes in synthetic time series. We then compare our tool to several others, showing that it achieves highest classification performance on a benchmark dataset–sequential MNIST–even when corrupted by noise. Additionally, we apply TRAKR to electrocorticography (ECoG) data from the macaque orbitofrontal cortex (OFC), a higher-order brain region involved in encoding the value of expected outcomes. We show that TRAKR can classify different behaviorally relevant epochs in the neural time series more accurately and efficiently than conventional approaches. Therefore, TRAKR can be used as a fast and accurate tool to distinguish patterns in complex nonlinear time-series data, such as neural recordings. | Reject |
ICLR.cc/2021/Conference | A Diffusion Theory For Deep Learning Dynamics: Stochastic Gradient Descent Exponentially Favors Flat Minima | Stochastic Gradient Descent (SGD) and its variants are mainstream methods for training deep networks in practice. SGD is known to find a flat minimum that often generalizes well. However, it is mathematically unclear how deep learning can select a flat minimum among so many minima. To answer the question quantitatively, we develop a density diffusion theory to reveal how minima selection quantitatively depends on the minima sharpness and the hyperparameters. To the best of our knowledge, we are the first to theoretically and empirically prove that, benefited from the Hessian-dependent covariance of stochastic gradient noise, SGD favors flat minima exponentially more than sharp minima, while Gradient Descent (GD) with injected white noise favors flat minima only polynomially more than sharp minima. We also reveal that either a small learning rate or large-batch training requires exponentially many iterations to escape from minima in terms of the ratio of the batch size and learning rate. Thus, large-batch training cannot search flat minima efficiently in a realistic computational time. | Accept (Poster) |
ICLR.cc/2023/Conference | UNICORN: A Unified Backdoor Trigger Inversion Framework | The backdoor attack, where the adversary uses inputs stamped with triggers (e.g., a patch) to activate pre-planted malicious behaviors, is a severe threat to Deep Neural Network (DNN) models. Trigger inversion is an effective way of identifying backdoor models and understanding embedded adversarial behaviors. A challenge of trigger inversion is that there are many ways of constructing the trigger. Existing methods cannot generalize to various types of triggers by making certain assumptions or attack-specific constraints. The fundamental reason is that existing work does not formally define the trigger and the inversion problem. This work formally defines and analyzes the trigger and the inversion problem. Then, it proposes a unified framework to invert backdoor triggers based on the formalization of triggers and the identified inner behaviors of backdoor models from our analysis. Our prototype UNICORN is general and effective in inverting backdoor triggers in DNNs. The code can be found at https://github.com/RU-System-Software-and-Security/UNICORN. | Accept: notable-top-25% |
ICLR.cc/2019/Conference | Lipschitz regularized Deep Neural Networks generalize | We show that if the usual training loss is augmented by a Lipschitz regularization term, then the networks generalize. We prove generalization by first establishing a stronger convergence result, along with a rate of convergence. A second result resolves a question posed in Zhang et al. (2016): how can a model distinguish between the case of clean labels, and randomized labels? Our answer is that Lipschitz regularization using the Lipschitz constant of the clean data makes this distinction. In this case, the model learns a different function which we hypothesize correctly fails to learn the dirty labels. | Reject |