Datasets:

CShorten

/

Last-Week-on-ML-ArXiv

like 0

When does Bias Transfer in Transfer Learning?

Using transfer learning to adapt a pre-trained "source model" to a downstream "target task" can dramatically increase performance with seemingly no downside. In this work, we demonstrate that there can exist a downside after all: bias transfer, or the tendency for biases of the source model to persist even after adapting the model to the target class. Through a combination of synthetic and natural experiments, we show that bias transfer both (a) arises in realistic settings (such as when pre-training on ImageNet or other standard datasets) and (b) can occur even when the target dataset is explicitly de-biased. As transfer-learned models are increasingly deployed in the real world, our work highlights the importance of understanding the limitations of pre-trained source models. Code is available at https://github.com/MadryLab/bias-transfer

Online Bilevel Optimization: Regret Analysis of Online Alternating Gradient Methods

Online optimization is a well-established optimization paradigm that aims to make a sequence of correct decisions given knowledge of the correct answer to previous decision tasks. Bilevel programming involves a hierarchical optimization problem where the feasible region of the so-called outer problem is restricted by the graph of the solution set mapping of the inner problem. This paper brings these two ideas together and studies an online bilevel optimization setting in which a sequence of time-varying bilevel problems are revealed one after the other. We extend the known regret bounds for single-level online algorithms to the bilevel setting. Specifically, we introduce new notions of bilevel regret, develop an online alternating time-averaged gradient method that is capable of leveraging smoothness, and provide regret bounds in terms of the path-length of the inner and outer minimizer sequences.

Strong Heuristics for Named Entity Linking

Named entity linking (NEL) in news is a challenging endeavour due to the frequency of unseen and emerging entities, which necessitates the use of unsupervised or zero-shot methods. However, such methods tend to come with caveats, such as no integration of suitable knowledge bases (like Wikidata) for emerging entities, a lack of scalability, and poor interpretability. Here, we consider person disambiguation in Quotebank, a massive corpus of speaker-attributed quotations from the news, and investigate the suitability of intuitive, lightweight, and scalable heuristics for NEL in web-scale corpora. Our best performing heuristic disambiguates 94% and 63% of the mentions on Quotebank and the AIDA-CoNLL benchmark, respectively. Additionally, the proposed heuristics compare favourably to the state-of-the-art unsupervised and zero-shot methods, Eigenthemes and mGENRE, respectively, thereby serving as strong baselines for unsupervised and zero-shot entity linking.

Improved conformalized quantile regression

Conformalized quantile regression is a procedure that inherits the advantages of conformal prediction and quantile regression. That is, we use quantile regression to estimate the true conditional quantile and then apply a conformal step on a calibration set to ensure marginal coverage. In this way, we get adaptive prediction intervals that account for heteroscedasticity. However, the aforementioned conformal step lacks adaptiveness as described in (Romano et al., 2019). To overcome this limitation, instead of applying a single conformal step after estimating conditional quantiles with quantile regression, we propose to cluster the explanatory variables weighted by their permutation importance with an optimized k-means and apply k conformal steps. To show that this improved version outperforms the classic version of conformalized quantile regression and is more adaptive to heteroscedasticity, we extensively compare the prediction intervals of both in open datasets.

A multi-task network approach for calculating discrimination-free insurance prices

In applications of predictive modeling, such as insurance pricing, indirect or proxy discrimination is an issue of major concern. Namely, there exists the possibility that protected policyholder characteristics are implicitly inferred from non-protected ones by predictive models, and are thus having an undesirable (or illegal) impact on prices. A technical solution to this problem relies on building a best-estimate model using all policyholder characteristics (including protected ones) and then averaging out the protected characteristics for calculating individual prices. However, such approaches require full knowledge of policyholders' protected characteristics, which may in itself be problematic. Here, we address this issue by using a multi-task neural network architecture for claim predictions, which can be trained using only partial information on protected characteristics, and it produces prices that are free from proxy discrimination. We demonstrate the use of the proposed model and we find that its predictive accuracy is comparable to a conventional feedforward neural network (on full information). However, this multi-task network has clearly superior performance in the case of partially missing policyholder information.

The Intrinsic Manifolds of Radiological Images and their Role in Deep Learning

The manifold hypothesis is a core mechanism behind the success of deep learning, so understanding the intrinsic manifold structure of image data is central to studying how neural networks learn from the data. Intrinsic dataset manifolds and their relationship to learning difficulty have recently begun to be studied for the common domain of natural images, but little such research has been attempted for radiological images. We address this here. First, we compare the intrinsic manifold dimensionality of radiological and natural images. We also investigate the relationship between intrinsic dimensionality and generalization ability over a wide range of datasets. Our analysis shows that natural image datasets generally have a higher number of intrinsic dimensions than radiological images. However, the relationship between generalization ability and intrinsic dimensionality is much stronger for medical images, which could be explained as radiological images having intrinsic features that are more difficult to learn. These results give a more principled underpinning for the intuition that radiological images can be more challenging to apply deep learning to than natural image datasets common to machine learning research. We believe rather than directly applying models developed for natural images to the radiological imaging domain, more care should be taken to developing architectures and algorithms that are more tailored to the specific characteristics of this domain. The research shown in our paper, demonstrating these characteristics and the differences from natural images, is an important first step in this direction.

Private Matrix Approximation and Geometry of Unitary Orbits

Consider the following optimization problem: Given $n \times n$ matrices $A$ and $\Lambda$, maximize $\langle A, U\Lambda U^*\rangle$ where $U$ varies over the unitary group $\mathrm{U}(n)$. This problem seeks to approximate $A$ by a matrix whose spectrum is the same as $\Lambda$ and, by setting $\Lambda$ to be appropriate diagonal matrices, one can recover matrix approximation problems such as PCA and rank-$k$ approximation. We study the problem of designing differentially private algorithms for this optimization problem in settings where the matrix $A$ is constructed using users' private data. We give efficient and private algorithms that come with upper and lower bounds on the approximation error. Our results unify and improve upon several prior works on private matrix approximation problems. They rely on extensions of packing/covering number bounds for Grassmannians to unitary orbits which should be of independent interest.

Astroconformer: Inferring Surface Gravity of Stars from Stellar Light Curves with Transformer

We introduce Astroconformer, a Transformer-based model to analyze stellar light curves from the Kepler mission. We demonstrate that Astrconformer can robustly infer the stellar surface gravity as a supervised task. Importantly, as Transformer captures long-range information in the time series, it outperforms the state-of-the-art data-driven method in the field, and the critical role of self-attention is proved through ablation experiments. Furthermore, the attention map from Astroconformer exemplifies the long-range correlation information learned by the model, leading to a more interpretable deep learning approach for asteroseismology. Besides data from Kepler, we also show that the method can generalize to sparse cadence light curves from the Rubin Observatory, paving the way for the new era of asteroseismology, harnessing information from long-cadence ground-based observations.

DIWIFT: Discovering Instance-wise Influential Features for Tabular Data

Tabular data is one of the most common data storage formats in business applications, ranging from retail, bank and E-commerce. These applications rely heavily on machine learning models to achieve business success. One of the critical problems in learning tabular data is to distinguish influential features from all the predetermined features. Global feature selection has been well-studied for quite some time, assuming that all instances have the same influential feature subsets. However, different instances rely on different feature subsets in practice, which also gives rise to that instance-wise feature selection receiving increasing attention in recent studies. In this paper, we first propose a novel method for discovering instance-wise influential features for tabular data (DIWIFT), the core of which is to introduce the influence function to measure the importance of an instance-wise feature. DIWIFT is capable of automatically discovering influential feature subsets of different sizes in different instances, which is different from global feature selection that considers all instances with the same influential feature subset. On the other hand, different from the previous instance-wise feature selection, DIWIFT minimizes the validation loss on the validation set and is thus more robust to the distribution shift existing in the training dataset and test dataset, which is important in tabular data. Finally, we conduct extensive experiments on both synthetic and real-world datasets to validate the effectiveness of our DIWIFT, compared it with baseline methods. Moreover, we also demonstrate the robustness of our method via some ablation experiments.

Enhancing Adversarial Attacks on Single-Layer NVM Crossbar-Based Neural Networks with Power Consumption Information

Adversarial attacks on state-of-the-art machine learning models pose a significant threat to the safety and security of mission-critical autonomous systems. This paper considers the additional vulnerability of machine learning models when attackers can measure the power consumption of their underlying hardware platform. In particular, we explore the utility of power consumption information for adversarial attacks on non-volatile memory crossbar-based single-layer neural networks. Our results from experiments with MNIST and CIFAR-10 datasets show that power consumption can reveal important information about the neural network's weight matrix, such as the 1-norm of its columns. That information can be used to infer the sensitivity of the network's loss with respect to different inputs. We also find that surrogate-based black box attacks that utilize crossbar power information can lead to improved attack efficiency.

BFE and AdaBFE: A New Approach in Learning Rate Automation for Stochastic Optimization

In this paper, a new gradient-based optimization approach by automatically adjusting the learning rate is proposed. This approach can be applied to design non-adaptive learning rate and adaptive learning rate. Firstly, I will introduce the non-adaptive learning rate optimization method: Binary Forward Exploration (BFE), and then the corresponding adaptive per-parameter learning rate method: Adaptive BFE (AdaBFE) is possible to be developed. This approach could be an alternative method to optimize the learning rate based on the stochastic gradient descent (SGD) algorithm besides the current non-adaptive learning rate methods e.g. SGD, momentum, Nesterov and the adaptive learning rate methods e.g. AdaGrad, AdaDelta, Adam... The purpose to develop this approach is not to beat the benchmark of other methods but just to provide a different perspective to optimize the gradient descent method, although some comparative study with previous methods will be made in the following sections. This approach is expected to be heuristic or inspire researchers to improve gradient-based optimization combined with previous methods.

Graph Trees with Attention

When dealing with tabular data, models based on regression and decision trees are a popular choice due to the high accuracy they provide on such tasks and their ease of application as compared to other model classes. Yet, when it comes to graph-structure data, current tree learning algorithms do not provide tools to manage the structure of the data other than relying on feature engineering. In this work we address the above gap, and introduce Graph Trees with Attention (GTA), a new family of tree-based learning algorithms that are designed to operate on graphs. GTA leverages both the graph structure and the features at the vertices and employs an attention mechanism that allows decisions to concentrate on sub-structures of the graph. We analyze GTA models and show that they are strictly more expressive than plain decision trees. We also demonstrate the benefits of GTA empirically on multiple graph and node prediction benchmarks. In these experiments, GTA always outperformed other tree-based models and often outperformed other types of graph-learning algorithms such as Graph Neural Networks (GNNs) and Graph Kernels. Finally, we also provide an explainability mechanism for GTA, and demonstrate it can provide intuitive explanations.

Robust Counterfactual Explanations for Tree-Based Ensembles

Counterfactual explanations inform ways to achieve a desired outcome from a machine learning model. However, such explanations are not robust to certain real-world changes in the underlying model (e.g., retraining the model, changing hyperparameters, etc.), questioning their reliability in several applications, e.g., credit lending. In this work, we propose a novel strategy -- that we call RobX -- to generate robust counterfactuals for tree-based ensembles, e.g., XGBoost. Tree-based ensembles pose additional challenges in robust counterfactual generation, e.g., they have a non-smooth and non-differentiable objective function, and they can change a lot in the parameter space under retraining on very similar data. We first introduce a novel metric -- that we call Counterfactual Stability -- that attempts to quantify how robust a counterfactual is going to be to model changes under retraining, and comes with desirable theoretical properties. Our proposed strategy RobX works with any counterfactual generation method (base method) and searches for robust counterfactuals by iteratively refining the counterfactual generated by the base method using our metric Counterfactual Stability. We compare the performance of RobX with popular counterfactual generation methods (for tree-based ensembles) across benchmark datasets. The results demonstrate that our strategy generates counterfactuals that are significantly more robust (nearly 100% validity after actual model changes) and also realistic (in terms of local outlier factor) over existing state-of-the-art methods.

A Hybrid Approach for Binary Classification of Imbalanced Data

Binary classification with an imbalanced dataset is challenging. Models tend to consider all samples as belonging to the majority class. Although existing solutions such as sampling methods, cost-sensitive methods, and ensemble learning methods improve the poor accuracy of the minority class, these methods are limited by overfitting problems or cost parameters that are difficult to decide. We propose HADR, a hybrid approach with dimension reduction that consists of data block construction, dimentionality reduction, and ensemble learning with deep neural network classifiers. We evaluate the performance on eight imbalanced public datasets in terms of recall, G-mean, and AUC. The results show that our model outperforms state-of-the-art methods.

Towards the Use of Saliency Maps for Explaining Low-Quality Electrocardiograms to End Users

When using medical images for diagnosis, either by clinicians or artificial intelligence (AI) systems, it is important that the images are of high quality. When an image is of low quality, the medical exam that produced the image often needs to be redone. In telemedicine, a common problem is that the quality issue is only flagged once the patient has left the clinic, meaning they must return in order to have the exam redone. This can be especially difficult for people living in remote regions, who make up a substantial portion of the patients at Portal Telemedicina, a digital healthcare organization based in Brazil. In this paper, we report on ongoing work regarding (i) the development of an AI system for flagging and explaining low-quality medical images in real-time, (ii) an interview study to understand the explanation needs of stakeholders using the AI system at OurCompany, and, (iii) a longitudinal user study design to examine the effect of including explanations on the workflow of the technicians in our clinics. To the best of our knowledge, this would be the first longitudinal study on evaluating the effects of XAI methods on end-users -- stakeholders that use AI systems but do not have AI-specific expertise. We welcome feedback and suggestions on our experimental setup.

Pre-training Transformers for Molecular Property Prediction Using Reaction Prediction

Molecular property prediction is essential in chemistry, especially for drug discovery applications. However, available molecular property data is often limited, encouraging the transfer of information from related data. Transfer learning has had a tremendous impact in fields like Computer Vision and Natural Language Processing signaling for its potential in molecular property prediction. We present a pre-training procedure for molecular representation learning using reaction data and use it to pre-train a SMILES Transformer. We fine-tune and evaluate the pre-trained model on 12 molecular property prediction tasks from MoleculeNet within physical chemistry, biophysics, and physiology and show a statistically significant positive effect on 5 of the 12 tasks compared to a non-pre-trained baseline model.

Variational Flow Graphical Model

This paper introduces a novel approach to embed flow-based models with hierarchical structures. The proposed framework is named Variational Flow Graphical (VFG) Model. VFGs learn the representation of high dimensional data via a message-passing scheme by integrating flow-based functions through variational inference. By leveraging the expressive power of neural networks, VFGs produce a representation of the data using a lower dimension, thus overcoming the drawbacks of many flow-based models, usually requiring a high dimensional latent space involving many trivial variables. Aggregation nodes are introduced in the VFG models to integrate forward-backward hierarchical information via a message passing scheme. Maximizing the evidence lower bound (ELBO) of data likelihood aligns the forward and backward messages in each aggregation node achieving a consistency node state. Algorithms have been developed to learn model parameters through gradient updating regarding the ELBO objective. The consistency of aggregation nodes enable VFGs to be applicable in tractable inference on graphical structures. Besides representation learning and numerical inference, VFGs provide a new approach for distribution modeling on datasets with graphical latent structures. Additionally, theoretical study shows that VFGs are universal approximators by leveraging the implicitly invertible flow-based structures. With flexible graphical structures and superior excessive power, VFGs could potentially be used to improve probabilistic inference. In the experiments, VFGs achieves improved evidence lower bound (ELBO) and likelihood values on multiple datasets.

Histopathology DatasetGAN: Synthesizing Large-Resolution Histopathology Datasets

Self-supervised learning (SSL) methods are enabling an increasing number of deep learning models to be trained on image datasets in domains where labels are difficult to obtain. These methods, however, struggle to scale to the high resolution of medical imaging datasets, where they are critical for achieving good generalization on label-scarce medical image datasets. In this work, we propose the Histopathology DatasetGAN (HDGAN) framework, an extension of the DatasetGAN semi-supervised framework for image generation and segmentation that scales well to large-resolution histopathology images. We make several adaptations from the original framework, including updating the generative backbone, selectively extracting latent features from the generator, and switching to memory-mapped arrays. These changes reduce the memory consumption of the framework, improving its applicability to medical imaging domains. We evaluate HDGAN on a thrombotic microangiopathy high-resolution tile dataset, demonstrating strong performance on the high-resolution image-annotation generation task. We hope that this work enables more application of deep learning models to medical datasets, in addition to encouraging more exploration of self-supervised frameworks within the medical imaging domain.

Scaling Private Deep Learning with Low-Rank and Sparse Gradients

Applying Differentially Private Stochastic Gradient Descent (DPSGD) to training modern, large-scale neural networks such as transformer-based models is a challenging task, as the magnitude of noise added to the gradients at each iteration scales with model dimension, hindering the learning capability significantly. We propose a unified framework, $\textsf{LSG}$, that fully exploits the low-rank and sparse structure of neural networks to reduce the dimension of gradient updates, and hence alleviate the negative impacts of DPSGD. The gradient updates are first approximated with a pair of low-rank matrices. Then, a novel strategy is utilized to sparsify the gradients, resulting in low-dimensional, less noisy updates that are yet capable of retaining the performance of neural networks. Empirical evaluation on natural language processing and computer vision tasks shows that our method outperforms other state-of-the-art baselines.

Effective and Efficient Training for Sequential Recommendation using Recency Sampling

Many modern sequential recommender systems use deep neural networks, which can effectively estimate the relevance of items but require a lot of time to train. Slow training increases expenses, hinders product development timescales and prevents the model from being regularly updated to adapt to changing user preferences. Training such sequential models involves appropriately sampling past user interactions to create a realistic training objective. The existing training objectives have limitations. For instance, next item prediction never uses the beginning of the sequence as a learning target, thereby potentially discarding valuable data. On the other hand, the item masking used by BERT4Rec is only weakly related to the goal of the sequential recommendation; therefore, it requires much more time to obtain an effective model. Hence, we propose a novel Recency-based Sampling of Sequences training objective that addresses both limitations. We apply our method to various recent and state-of-the-art model architectures - such as GRU4Rec, Caser, and SASRec. We show that the models enhanced with our method can achieve performances exceeding or very close to stateof-the-art BERT4Rec, but with much less training time.

When does SGD favor flat minima? A quantitative characterization via linear stability

The observation that stochastic gradient descent (SGD) favors flat minima has played a fundamental role in understanding implicit regularization of SGD and guiding the tuning of hyperparameters. In this paper, we provide a quantitative explanation of this striking phenomenon by relating the particular noise structure of SGD to its \emph{linear stability} (Wu et al., 2018). Specifically, we consider training over-parameterized models with square loss. We prove that if a global minimum $\theta^*$ is linearly stable for SGD, then it must satisfy $\|H(\theta^*)\|_F\leq O(\sqrt{B}/\eta)$, where $\|H(\theta^*)\|_F, B,\eta$ denote the Frobenius norm of Hessian at $\theta^*$, batch size, and learning rate, respectively. Otherwise, SGD will escape from that minimum \emph{exponentially} fast. Hence, for minima accessible to SGD, the flatness -- as measured by the Frobenius norm of the Hessian -- is bounded independently of the model size and sample size. The key to obtaining these results is exploiting the particular geometry awareness of SGD noise: 1) the noise magnitude is proportional to loss value; 2) the noise directions concentrate in the sharp directions of local landscape. This property of SGD noise provably holds for linear networks and random feature models (RFMs) and is empirically verified for nonlinear networks. Moreover, the validity and practical relevance of our theoretical findings are justified by extensive numerical experiments.

Predicting is not Understanding: Recognizing and Addressing Underspecification in Machine Learning

Machine learning (ML) models are typically optimized for their accuracy on a given dataset. However, this predictive criterion rarely captures all desirable properties of a model, in particular how well it matches a domain expert's understanding of a task. Underspecification refers to the existence of multiple models that are indistinguishable in their in-domain accuracy, even though they differ in other desirable properties such as out-of-distribution (OOD) performance. Identifying these situations is critical for assessing the reliability of ML models. We formalize the concept of underspecification and propose a method to identify and partially address it. We train multiple models with an independence constraint that forces them to implement different functions. They discover predictive features that are otherwise ignored by standard empirical risk minimization (ERM), which we then distill into a global model with superior OOD performance. Importantly, we constrain the models to align with the data manifold to ensure that they discover meaningful features. We demonstrate the method on multiple datasets in computer vision (collages, WILDS-Camelyon17, GQA) and discuss general implications of underspecification. Most notably, in-domain performance cannot serve for OOD model selection without additional assumptions.

Cascaded Deep Hybrid Models for Multistep Household Energy Consumption Forecasting

Sustainability requires increased energy efficiency with minimal waste. The future power systems should thus provide high levels of flexibility iin controling energy consumption. Precise projections of future energy demand/load at the aggregate and on the individual site levels are of great importance for decision makers and professionals in the energy industry. Forecasting energy loads has become more advantageous for energy providers and customers, allowing them to establish an efficient production strategy to satisfy demand. This study introduces two hybrid cascaded models for forecasting multistep household power consumption in different resolutions. The first model integrates Stationary Wavelet Transform (SWT), as an efficient signal preprocessing technique, with Convolutional Neural Networks and Long Short Term Memory (LSTM). The second hybrid model combines SWT with a self-attention based neural network architecture named transformer. The major constraint of using time-frequency analysis methods such as SWT in multistep energy forecasting problems is that they require sequential signals, making signal reconstruction problematic in multistep forecasting applications.The cascaded models can efficiently address this problem through using the recursive outputs. Experimental results show that the proposed hybrid models achieve superior prediction performance compared to the existing multistep power consumption prediction methods. The results will pave the way for more accurate and reliable forecasting of household power consumption.

Instance-Dependent Near-Optimal Policy Identification in Linear MDPs via Online Experiment Design

While much progress has been made in understanding the minimax sample complexity of reinforcement learning (RL) -- the complexity of learning on the "worst-case" instance -- such measures of complexity often do not capture the true difficulty of learning. In practice, on an "easy" instance, we might hope to achieve a complexity far better than that achievable on the worst-case instance. In this work we seek to understand the "instance-dependent" complexity of learning near-optimal policies (PAC RL) in the setting of RL with linear function approximation. We propose an algorithm, \textsc{Pedel}, which achieves a fine-grained instance-dependent measure of complexity, the first of its kind in the RL with function approximation setting, thereby capturing the difficulty of learning on each particular problem instance. Through an explicit example, we show that \textsc{Pedel} yields provable gains over low-regret, minimax-optimal algorithms and that such algorithms are unable to hit the instance-optimal rate. Our approach relies on a novel online experiment design-based procedure which focuses the exploration budget on the "directions" most relevant to learning a near-optimal policy, and may be of independent interest.

voxel2vec: A Natural Language Processing Approach to Learning Distributed Representations for Scientific Data

Relationships in scientific data, such as the numerical and spatial distribution relations of features in univariate data, the scalar-value combinations' relations in multivariate data, and the association of volumes in time-varying and ensemble data, are intricate and complex. This paper presents voxel2vec, a novel unsupervised representation learning model, which is used to learn distributed representations of scalar values/scalar-value combinations in a low-dimensional vector space. Its basic assumption is that if two scalar values/scalar-value combinations have similar contexts, they usually have high similarity in terms of features. By representing scalar values/scalar-value combinations as symbols, voxel2vec learns the similarity between them in the context of spatial distribution and then allows us to explore the overall association between volumes by transfer prediction. We demonstrate the usefulness and effectiveness of voxel2vec by comparing it with the isosurface similarity map of univariate data and applying the learned distributed representations to feature classification for multivariate data and to association analysis for time-varying and ensemble data.

Simple and Efficient Heterogeneous Graph Neural Network

Heterogeneous graph neural networks (HGNNs) deliver the powerful capability to embed rich structural and semantic information of a heterogeneous graph into low-dimensional node representations. Existing HGNNs usually learn to embed information using hierarchy attention mechanism and repeated neighbor aggregation, suffering from unnecessary complexity and redundant computation. This paper proposes Simple and Efficient Heterogeneous Graph Neural Network (SeHGNN) which reduces this excess complexity through avoiding overused node-level attention within the same relation and pre-computing the neighbor aggregation in the pre-processing stage. Unlike previous work, SeHGNN utilizes a light-weight parameter-free neighbor aggregator to learn structural information for each metapath, and a transformer-based semantic aggregator to combine semantic information across metapaths for the final embedding of each node. As a result, SeHGNN offers the simple network structure, high prediction accuracy, and fast training speed. Extensive experiments on five real-world heterogeneous graphs demonstrate the superiority of SeHGNN over the state-of-the-arts on both the accuracy and training speed. Codes are available at https://github.com/ICT-GIMLab/SeHGNN.

AI-enhanced iterative solvers for accelerating the solution of large scale parametrized linear systems of equations

Recent advances in the field of machine learning open a new era in high performance computing. Applications of machine learning algorithms for the development of accurate and cost-efficient surrogates of complex problems have already attracted major attention from scientists. Despite their powerful approximation capabilities, however, surrogates cannot produce the `exact' solution to the problem. To address this issue, this paper exploits up-to-date ML tools and delivers customized iterative solvers of linear equation systems, capable of solving large-scale parametrized problems at any desired level of accuracy. Specifically, the proposed approach consists of the following two steps. At first, a reduced set of model evaluations is performed and the corresponding solutions are used to establish an approximate mapping from the problem's parametric space to its solution space using deep feedforward neural networks and convolutional autoencoders. This mapping serves a means to obtain very accurate initial predictions of the system's response to new query points at negligible computational cost. Subsequently, an iterative solver inspired by the Algebraic Multigrid method in combination with Proper Orthogonal Decomposition, termed POD-2G, is developed that successively refines the initial predictions towards the exact system solutions. The application of POD-2G as a standalone solver or as preconditioner in the context of preconditioned conjugate gradient methods is demonstrated on several numerical examples of large scale systems, with the results indicating its superiority over conventional iterative solution schemes.

Tractable Dendritic RNNs for Reconstructing Nonlinear Dynamical Systems

In many scientific disciplines, we are interested in inferring the nonlinear dynamical system underlying a set of observed time series, a challenging task in the face of chaotic behavior and noise. Previous deep learning approaches toward this goal often suffered from a lack of interpretability and tractability. In particular, the high-dimensional latent spaces often required for a faithful embedding, even when the underlying dynamics lives on a lower-dimensional manifold, can hamper theoretical analysis. Motivated by the emerging principles of dendritic computation, we augment a dynamically interpretable and mathematically tractable piecewise-linear (PL) recurrent neural network (RNN) by a linear spline basis expansion. We show that this approach retains all the theoretically appealing properties of the simple PLRNN, yet boosts its capacity for approximating arbitrary nonlinear dynamical systems in comparatively low dimensions. We employ two frameworks for training the system, one combining back-propagation-through-time (BPTT) with teacher forcing, and another based on fast and scalable variational inference. We show that the dendritically expanded PLRNN achieves better reconstructions with fewer parameters and dimensions on various dynamical systems benchmarks and compares favorably to other methods, while retaining a tractable and interpretable structure.

Transformers discover an elementary calculation system exploiting local attention and grid-like problem representation

Mathematical reasoning is one of the most impressive achievements of human intellect but remains a formidable challenge for artificial intelligence systems. In this work we explore whether modern deep learning architectures can learn to solve a symbolic addition task by discovering effective arithmetic procedures. Although the problem might seem trivial at first glance, generalizing arithmetic knowledge to operations involving a higher number of terms, possibly composed by longer sequences of digits, has proven extremely challenging for neural networks. Here we show that universal transformers equipped with local attention and adaptive halting mechanisms can learn to exploit an external, grid-like memory to carry out multi-digit addition. The proposed model achieves remarkable accuracy even when tested with problems requiring extrapolation outside the training distribution; most notably, it does so by discovering human-like calculation strategies such as place value alignment.

Compositional Generalization in Grounded Language Learning via Induced Model Sparsity

We provide a study of how induced model sparsity can help achieve compositional generalization and better sample efficiency in grounded language learning problems. We consider simple language-conditioned navigation problems in a grid world environment with disentangled observations. We show that standard neural architectures do not always yield compositional generalization. To address this, we design an agent that contains a goal identification module that encourages sparse correlations between words in the instruction and attributes of objects, composing them together to find the goal. The output of the goal identification module is the input to a value iteration network planner. Our agent maintains a high level of performance on goals containing novel combinations of properties even when learning from a handful of demonstrations. We examine the internal representations of our agent and find the correct correspondences between words in its dictionary and attributes in the environment.

Pure Transformers are Powerful Graph Learners

We show that standard Transformers without graph-specific modifications can lead to promising results in graph learning both in theory and practice. Given a graph, we simply treat all nodes and edges as independent tokens, augment them with token embeddings, and feed them to a Transformer. With an appropriate choice of token embeddings, we prove that this approach is theoretically at least as expressive as an invariant graph network (2-IGN) composed of equivariant linear layers, which is already more expressive than all message-passing Graph Neural Networks (GNN). When trained on a large-scale graph dataset (PCQM4Mv2), our method coined Tokenized Graph Transformer (TokenGT) achieves significantly better results compared to GNN baselines and competitive results compared to Transformer variants with sophisticated graph-specific inductive bias. Our implementation is available at https://github.com/jw9730/tokengt.

Quantitative Assessment of DESIS Hyperspectral Data for Plant Biodiversity Estimation in Australia

Diversity of terrestrial plants plays a key role in maintaining a stable, healthy, and productive ecosystem. Though remote sensing has been seen as a promising and cost-effective proxy for estimating plant diversity, there is a lack of quantitative studies on how confidently plant diversity can be inferred from spaceborne hyperspectral data. In this study, we assessed the ability of hyperspectral data captured by the DLR Earth Sensing Imaging Spectrometer (DESIS) for estimating plant species richness in the Southern Tablelands and Snowy Mountains regions in southeast Australia. Spectral features were firstly extracted from DESIS spectra with principal component analysis, canonical correlation analysis, and partial least squares analysis. Then regression was conducted between the extracted features and plant species richness with ordinary least squares regression, kernel ridge regression, and Gaussian process regression. Results were assessed with the coefficient of correlation ($r$) and Root-Mean-Square Error (RMSE), based on a two-fold cross validation scheme. With the best performing model, $r$ is 0.71 and RMSE is 5.99 for the Southern Tablelands region, while $r$ is 0.62 and RMSE is 6.20 for the Snowy Mountains region. The assessment results reported in this study provide supports for future studies on understanding the relationship between spaceborne hyperspectral measurements and terrestrial plant biodiversity.

Multi-Contrast MRI Segmentation Trained on Synthetic Images

In our comprehensive experiments and evaluations, we show that it is possible to generate multiple contrast (even all synthetically) and use synthetically generated images to train an image segmentation engine. We showed promising segmentation results tested on real multi-contrast MRI scans when delineating muscle, fat, bone and bone marrow, all trained on synthetic images. Based on synthetic image training, our segmentation results were as high as 93.91\%, 94.11\%, 91.63\%, 95.33\%, for muscle, fat, bone, and bone marrow delineation, respectively. Results were not significantly different from the ones obtained when real images were used for segmentation training: 94.68\%, 94.67\%, 95.91\%, and 96.82\%, respectively.

Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives

Ordinal regression with anchored reference samples (ORARS) has been proposed for predicting the subjective Mean Opinion Score (MOS) of input stimuli automatically. The ORARS addresses the MOS prediction problem by pairing a test sample with each of the pre-scored anchored reference samples. A trained binary classifier is then used to predict which sample, test or anchor, is better statistically. Posteriors of the binary preference decision are then used to predict the MOS of the test sample. In this paper, rigorous framework, analysis, and experiments to demonstrate that ORARS are advantageous over simple regressions are presented. The contributions of this work are: 1) Show that traditional regression can be reformulated into multiple preference tests to yield a better performance, which is confirmed with simulations experimentally; 2) Generalize ORARS to other regression problems and verify its effectiveness; 3) Provide some prerequisite conditions which can insure proper application of ORARS.

Information Compression and Performance Evaluation of Tic-Tac-Toe's Evaluation Function Using Singular Value Decomposition

We approximated the evaluation function for the game Tic-Tac-Toe by singular value decomposition (SVD) and investigated the effect of approximation accuracy on winning rate. We first prepared the perfect evaluation function of Tic-Tac-Toe and performed low-rank approximation by considering the evaluation function as a ninth-order tensor. We found that we can reduce the amount of information of the evaluation function by 70% without significantly degrading the performance. Approximation accuracy and winning rate were strongly correlated but not perfectly proportional. We also investigated how the decomposition method of the evaluation function affects the performance. We considered two decomposition methods: simple SVD regarding the evaluation function as a matrix and the Tucker decomposition by higher-order SVD (HOSVD). At the same compression ratio, the strategy with the approximated evaluation function obtained by HOSVD exhibited a significantly higher winning rate than that obtained by SVD. These results suggest that SVD can effectively compress board game strategies and an optimal compression method that depends on the game exists.

Nonparametric Factor Trajectory Learning for Dynamic Tensor Decomposition

Tensor decomposition is a fundamental framework to analyze data that can be represented by multi-dimensional arrays. In practice, tensor data is often accompanied by temporal information, namely the time points when the entry values were generated. This information implies abundant, complex temporal variation patterns. However, current methods always assume the factor representations of the entities in each tensor mode are static, and never consider their temporal evolution. To fill this gap, we propose NONparametric FActor Trajectory learning for dynamic tensor decomposition (NONFAT). We place Gaussian process (GP) priors in the frequency domain and conduct inverse Fourier transform via Gauss-Laguerre quadrature to sample the trajectory functions. In this way, we can overcome data sparsity and obtain robust trajectory estimates across long time horizons. Given the trajectory values at specific time points, we use a second-level GP to sample the entry values and to capture the temporal relationship between the entities. For efficient and scalable inference, we leverage the matrix Gaussian structure in the model, introduce a matrix Gaussian posterior, and develop a nested sparse variational learning algorithm. We have shown the advantage of our method in several real-world applications.

Distillation to Enhance the Portability of Risk Models Across Institutions with Large Patient Claims Database

Artificial intelligence, and particularly machine learning (ML), is increasingly developed and deployed to support healthcare in a variety of settings. However, clinical decision support (CDS) technologies based on ML need to be portable if they are to be adopted on a broad scale. In this respect, models developed at one institution should be reusable at another. Yet there are numerous examples of portability failure, particularly due to naive application of ML models. Portability failure can lead to suboptimal care and medical errors, which ultimately could prevent the adoption of ML-based CDS in practice. One specific healthcare challenge that could benefit from enhanced portability is the prediction of 30-day readmission risk. Research to date has shown that deep learning models can be effective at modeling such risk. In this work, we investigate the practicality of model portability through a cross-site evaluation of readmission prediction models. To do so, we apply a recurrent neural network, augmented with self-attention and blended with expert features, to build readmission prediction models for two independent large scale claims datasets. We further present a novel transfer learning technique that adapts the well-known method of born-again network (BAN) training. Our experiments show that direct application of ML models trained at one institution and tested at another institution perform worse than models trained and tested at the same institution. We further show that the transfer learning approach based on the BAN produces models that are better than those trained on just a single institution's data. Notably, this improvement is consistent across both sites and occurs after a single retraining, which illustrates the potential for a cheap and general model transfer mechanism of readmission risk prediction.

Transformers are Adaptable Task Planners

Every home is different, and every person likes things done in their particular way. Therefore, home robots of the future need to both reason about the sequential nature of day-to-day tasks and generalize to user's preferences. To this end, we propose a Transformer Task Planner(TTP) that learns high-level actions from demonstrations by leveraging object attribute-based representations. TTP can be pre-trained on multiple preferences and shows generalization to unseen preferences using a single demonstration as a prompt in a simulated dishwasher loading task. Further, we demonstrate real-world dish rearrangement using TTP with a Franka Panda robotic arm, prompted using a single human demonstration.

PAC Prediction Sets for Meta-Learning

Uncertainty quantification is a key component of machine learning models targeted at safety-critical systems such as in healthcare or autonomous vehicles. We study this problem in the context of meta learning, where the goal is to quickly adapt a predictor to new tasks. In particular, we propose a novel algorithm to construct \emph{PAC prediction sets}, which capture uncertainty via sets of labels, that can be adapted to new tasks with only a few training examples. These prediction sets satisfy an extension of the typical PAC guarantee to the meta learning setting; in particular, the PAC guarantee holds with high probability over future tasks. We demonstrate the efficacy of our approach on four datasets across three application domains: mini-ImageNet and CIFAR10-C in the visual domain, FewRel in the language domain, and the CDC Heart Dataset in the medical domain. In particular, our prediction sets satisfy the PAC guarantee while having smaller size compared to other baselines that also satisfy this guarantee.

EEPT: Early Discovery of Emerging Entities in Twitter with Semantic Similarity

Some events which happen in the future could be important for companies, governments, and even our personal life. Prediction of these events before their establishment is helpful for efficient decision-making. We call such events emerging entities. They have not taken place yet, and there is no information about them in KB. However, some clues exist in different areas, especially on social media. Thus, retrieving these type of entities are possible. This paper proposes a method of early discovery of emerging entities. We use semantic clustering of short messages. To evaluate the performance of our proposal, we devise and utilize a performance evaluation metric. The results show that our proposed method finds those emerging entities of which Twitter trends are not always capable.

GAMa: Cross-view Video Geo-localization

The existing work in cross-view geo-localization is based on images where a ground panorama is matched to an aerial image. In this work, we focus on ground videos instead of images which provides additional contextual cues which are important for this task. There are no existing datasets for this problem, therefore we propose GAMa dataset, a large-scale dataset with ground videos and corresponding aerial images. We also propose a novel approach to solve this problem. At clip-level, a short video clip is matched with corresponding aerial image and is later used to get video-level geo-localization of a long video. Moreover, we propose a hierarchical approach to further improve the clip-level geolocalization. It is a challenging dataset, unaligned and limited field of view, and our proposed method achieves a Top-1 recall rate of 19.4% and 45.1% @1.0mile. Code and dataset are available at following link: https://github.com/svyas23/GAMa.

Composite FORCE learning of chaotic echo state networks for time-series prediction

Echo state network (ESN), a kind of recurrent neural networks, consists of a fixed reservoir in which neurons are connected randomly and recursively and obtains the desired output only by training output connection weights. First-order reduced and controlled error (FORCE) learning is an online supervised training approach that can change the chaotic activity of ESNs into specified activity patterns. This paper proposes a composite FORCE learning method based on recursive least squares to train ESNs whose initial activity is spontaneously chaotic, where a composite learning technique featured by dynamic regressor extension and memory data exploitation is applied to enhance parameter convergence. The proposed method is applied to a benchmark problem about predicting chaotic time series generated by the Mackey-Glass system, and numerical results have shown that it significantly improves learning and prediction performances compared with existing methods.

BioTABQA: Instruction Learning for Biomedical Table Question Answering

Table Question Answering (TQA) is an important but under-explored task. Most of the existing QA datasets are in unstructured text format and only few of them use tables as the context. To the best of our knowledge, none of TQA datasets exist in the biomedical domain where tables are frequently used to present information. In this paper, we first curate a table question answering dataset, BioTABQA, using 22 templates and the context from a biomedical textbook on differential diagnosis. BioTABQA can not only be used to teach a model how to answer questions from tables but also evaluate how a model generalizes to unseen questions, an important scenario for biomedical applications. To achieve the generalization evaluation, we divide the templates into 17 training and 5 cross-task evaluations. Then, we develop two baselines using single and multi-tasks learning on BioTABQA. Furthermore, we explore instructional learning, a recent technique showing impressive generalizing performance. Experimental results show that our instruction-tuned model outperforms single and multi-task baselines on an average by ~23% and ~6% across various evaluation settings, and more importantly, instruction-tuned model outperforms baselines by ~5% on cross-tasks.

A Deep Model for Partial Multi-Label Image Classification with Curriculum Based Disambiguation

In this paper, we study the partial multi-label (PML) image classification problem, where each image is annotated with a candidate label set consists of multiple relevant labels and other noisy labels. Existing PML methods typically design a disambiguation strategy to filter out noisy labels by utilizing prior knowledge with extra assumptions, which unfortunately is unavailable in many real tasks. Furthermore, because the objective function for disambiguation is usually elaborately designed on the whole training set, it can be hardly optimized in a deep model with SGD on mini-batches. In this paper, for the first time we propose a deep model for PML to enhance the representation and discrimination ability. On one hand, we propose a novel curriculum based disambiguation strategy to progressively identify ground-truth labels by incorporating the varied difficulties of different classes. On the other hand, a consistency regularization is introduced for model retraining to balance fitting identified easy labels and exploiting potential relevant labels. Extensive experimental results on the commonly used benchmark datasets show the proposed method significantly outperforms the SOTA methods.

Careful seeding for the k-medoids algorithm with incremental k++ cluster construction

The k-medoids algorithm is a popular variant of the k-means algorithm and widely used in pattern recognition and machine learning. A main drawback of the k-medoids algorithm is that it can be trapped in local optima. An improved k-medoids algorithm (INCKM) was recently proposed to overcome this drawback, based on constructing a candidate medoids subset with a parameter choosing procedure, but it may fail when dealing with imbalanced datasets. In this paper, we propose a novel incremental k-medoids algorithm (INCKPP) which dynamically increases the number of clusters from 2 to k through a nonparametric and stochastic k-means++ search procedure. Our algorithm can overcome the parameter selection problem in the improved k-medoids algorithm, improve the clustering performance, and deal with imbalanced datasets very well. But our algorithm has a weakness in computation efficiency. To address this issue, we propose a fast INCKPP algorithm (called INCKPP$_{sample}$) which preserves the computational efficiency of the simple and fast k-medoids algorithm with an improved clustering performance. The proposed algorithm is compared with three state-of-the-art algorithms: the improved k-medoids algorithm (INCKM), the simple and fast k-medoids algorithm (FKM) and the k-means++ algorithm (KPP). Extensive experiments on both synthetic and real world datasets including imbalanced datasets illustrate the effectiveness of the proposed algorithm.

Ensemble feature selection with clustering for analysis of high-dimensional, correlated clinical data in the search for Alzheimer's disease biomarkers

Healthcare datasets often contain groups of highly correlated features, such as features from the same biological system. When feature selection is applied to these datasets to identify the most important features, the biases inherent in some multivariate feature selectors due to correlated features make it difficult for these methods to distinguish between the important and irrelevant features and the results of the feature selection process can be unstable. Feature selection ensembles, which aggregate the results of multiple individual base feature selectors, have been investigated as a means of stabilising feature selection results, but do not address the problem of correlated features. We present a novel framework to create feature selection ensembles from multivariate feature selectors while taking into account the biases produced by groups of correlated features, using agglomerative hierarchical clustering in a pre-processing step. These methods were applied to two real-world datasets from studies of Alzheimer's disease (AD), a progressive neurodegenerative disease that has no cure and is not yet fully understood. Our results show a marked improvement in the stability of features selected over the models without clustering, and the features selected by these models are in keeping with the findings in the AD literature.

Text Enriched Sparse Hyperbolic Graph Convolutional Networks

Heterogeneous networks, which connect informative nodes containing text with different edge types, are routinely used to store and process information in various real-world applications. Graph Neural Networks (GNNs) and their hyperbolic variants provide a promising approach to encode such networks in a low-dimensional latent space through neighborhood aggregation and hierarchical feature extraction, respectively. However, these approaches typically ignore metapath structures and the available semantic information. Furthermore, these approaches are sensitive to the noise present in the training data. To tackle these limitations, in this paper, we propose Text Enriched Sparse Hyperbolic Graph Convolution Network (TESH-GCN) to capture the graph's metapath structures using semantic signals and further improve prediction in large heterogeneous graphs. In TESH-GCN, we extract semantic node information, which successively acts as a connection signal to extract relevant nodes' local neighborhood and graph-level metapath features from the sparse adjacency tensor in a reformulated hyperbolic graph convolution layer. These extracted features in conjunction with semantic features from the language model (for robustness) are used for the final downstream task. Experiments on various heterogeneous graph datasets show that our model outperforms the current state-of-the-art approaches by a large margin on the task of link prediction. We also report a reduction in both the training time and model parameters compared to the existing hyperbolic approaches through a reformulated hyperbolic graph convolution. Furthermore, we illustrate the robustness of our model by experimenting with different levels of simulated noise in both the graph structure and text, and also, present a mechanism to explain TESH-GCN's prediction by analyzing the extracted metapaths.

Linear Jamming Bandits: Sample-Efficient Learning for Non-Coherent Digital Jamming

It has been shown (Amuru et al. 2015) that online learning algorithms can be effectively used to select optimal physical layer parameters for jamming against digital modulation schemes without a priori knowledge of the victim's transmission strategy. However, this learning problem involves solving a multi-armed bandit problem with a mixed action space that can grow very large. As a result, convergence to the optimal jamming strategy can be slow, especially when the victim and jammer's symbols are not perfectly synchronized. In this work, we remedy the sample efficiency issues by introducing a linear bandit algorithm that accounts for inherent similarities between actions. Further, we propose context features which are well-suited for the statistical features of the non-coherent jamming problem and demonstrate significantly improved convergence behavior compared to the prior art. Additionally, we show how prior knowledge about the victim's transmissions can be seamlessly integrated into the learning framework. We finally discuss limitations in the asymptotic regime.

Instance-optimal PAC Algorithms for Contextual Bandits

In the stochastic contextual bandit setting, regret-minimizing algorithms have been extensively researched, but their instance-minimizing best-arm identification counterparts remain seldom studied. In this work, we focus on the stochastic bandit problem in the $(\epsilon,\delta)$-$\textit{PAC}$ setting: given a policy class $\Pi$ the goal of the learner is to return a policy $\pi\in \Pi$ whose expected reward is within $\epsilon$ of the optimal policy with probability greater than $1-\delta$. We characterize the first $\textit{instance-dependent}$ PAC sample complexity of contextual bandits through a quantity $\rho_{\Pi}$, and provide matching upper and lower bounds in terms of $\rho_{\Pi}$ for the agnostic and linear contextual best-arm identification settings. We show that no algorithm can be simultaneously minimax-optimal for regret minimization and instance-dependent PAC for best-arm identification. Our main result is a new instance-optimal and computationally efficient algorithm that relies on a polynomial number of calls to an argmax oracle.

Generalization to translation shifts: a study in architectures and augmentations

We provide a detailed evaluation of various image classification architectures (convolutional, vision transformer, and fully connected MLP networks) and data augmentation techniques towards generalization to large spacial translation shifts. We make the following observations: (a) In the absence of data augmentation, all architectures, including convolutional networks suffer degradation in performance when evaluated on translated test distributions. Understandably, both the in-distribution accuracy as well as degradation to shifts is significantly worse for non-convolutional architectures. (b) Across all architectures, even a minimal augmentation of $4$ pixel random crop improves the robustness of performance to much larger magnitude shifts of up to $1/4$ of image size ($8$-$16$ pixels) in the test data -- suggesting a form of meta generalization from augmentation. For non-convolutional architectures, while the absolute accuracy is still low, we see dramatic improvements in robustness to large translation shifts. (c) With sufficiently advanced augmentation ($4$ pixel crop+RandAugmentation+Erasing+MixUp) pipeline all architectures can be trained to have competitive performance, both in terms of in-distribution accuracy as well as generalization to large translation shifts.

Many-body localized hidden Born machine

Born Machines are quantum-inspired generative models that leverage the probabilistic nature of quantum states. Here, we present a new architecture called many-body localized (MBL) hidden Born machine that uses both MBL dynamics and hidden units as learning resources. We theoretically prove that MBL Born machines possess more expressive power than classical models, and the introduction of hidden units boosts its learning power. We numerically demonstrate that the MBL hidden Born machine is capable of learning a toy dataset consisting of patterns of MNIST handwritten digits, quantum data obtained from quantum many-body states, and non-local parity data. In order to understand the mechanism behind learning, we track physical quantities such as von Neumann entanglement entropy and Hamming distance during learning, and compare the learning outcomes in the MBL, thermal, and Anderson localized phases. We show that the superior learning power of the MBL phase relies importantly on both localization and interaction. Our architecture and algorithm provide novel strategies of utilizing quantum many-body systems as learning resources, and reveal a powerful connection between disorder, interaction, and learning in quantum systems.

Rethinking the Importance of Sampling in Physics-informed Neural Networks

Physics-informed neural networks (PINNs) have emerged as a powerful tool for solving partial differential equations (PDEs) in a variety of domains. While previous research in PINNs has mainly focused on constructing and balancing loss functions during training to avoid poor minima, the effect of sampling collocation points on the performance of PINNs has largely been overlooked. In this work, we find that the performance of PINNs can vary significantly with different sampling strategies, and using a fixed set of collocation points can be quite detrimental to the convergence of PINNs to the correct solution. In particular, (1) we hypothesize that training of PINNs rely on successful "propagation" of solution from initial and/or boundary condition points to interior points, and PINNs with poor sampling strategies can get stuck at trivial solutions if there are \textit{propagation failures}. (2) We demonstrate that propagation failures are characterized by highly imbalanced PDE residual fields where very high residuals are observed over very narrow regions. (3) To mitigate propagation failure, we propose a novel \textit{evolutionary sampling} (Evo) method that can incrementally accumulate collocation points in regions of high PDE residuals. We further provide an extension of Evo to respect the principle of causality while solving time-dependent PDEs. We empirically demonstrate the efficacy and efficiency of our proposed methods in a variety of PDE problems.

Federated and Transfer Learning: A Survey on Adversaries and Defense Mechanisms

The advent of federated learning has facilitated large-scale data exchange amongst machine learning models while maintaining privacy. Despite its brief history, federated learning is rapidly evolving to make wider use more practical. One of the most significant advancements in this domain is the incorporation of transfer learning into federated learning, which overcomes fundamental constraints of primary federated learning, particularly in terms of security. This chapter performs a comprehensive survey on the intersection of federated and transfer learning from a security point of view. The main goal of this study is to uncover potential vulnerabilities and defense mechanisms that might compromise the privacy and performance of systems that use federated and transfer learning.

Multi-Label Retinal Disease Classification using Transformers

Early detection of retinal diseases is one of the most important means of preventing partial or permanent blindness in patients. In this research, a novel multi-label classification system is proposed for the detection of multiple retinal diseases, using fundus images collected from a variety of sources. First, a new multi-label retinal disease dataset, the MuReD dataset, is constructed, using a number of publicly available datasets for fundus disease classification. Next, a sequence of post-processing steps is applied to ensure the quality of the image data and the range of diseases, present in the dataset. For the first time in fundus multi-label disease classification, a transformer-based model optimized through extensive experimentation is used for image analysis and decision making. Numerous experiments are performed to optimize the configuration of the proposed system. It is shown that the approach performs better than state-of-the-art works on the same task by 7.9% and 8.1% in terms of AUC score for disease detection and disease classification, respectively. The obtained results further support the potential applications of transformer-based architectures in the medical imaging field.

Unified Embeddings of Structural and Functional Connectome via a Function-Constrained Structural Graph Variational Auto-Encoder

Graph theoretical analyses have become standard tools in modeling functional and anatomical connectivity in the brain. With the advent of connectomics, the primary graphs or networks of interest are structural connectome (derived from DTI tractography) and functional connectome (derived from resting-state fMRI). However, most published connectome studies have focused on either structural or functional connectome, yet complementary information between them, when available in the same dataset, can be jointly leveraged to improve our understanding of the brain. To this end, we propose a function-constrained structural graph variational autoencoder (FCS-GVAE) capable of incorporating information from both functional and structural connectome in an unsupervised fashion. This leads to a joint low-dimensional embedding that establishes a unified spatial coordinate system for comparing across different subjects. We evaluate our approach using the publicly available OASIS-3 Alzheimer's disease (AD) dataset and show that a variational formulation is necessary to optimally encode functional brain dynamics. Further, the proposed joint embedding approach can more accurately distinguish different patient sub-populations than approaches that do not use complementary connectome information.

TractoFormer: A Novel Fiber-level Whole Brain Tractography Analysis Framework Using Spectral Embedding and Vision Transformers

Diffusion MRI tractography is an advanced imaging technique for quantitative mapping of the brain's structural connectivity. Whole brain tractography (WBT) data contains over hundreds of thousands of individual fiber streamlines (estimated brain connections), and this data is usually parcellated to create compact representations for data analysis applications such as disease classification. In this paper, we propose a novel parcellation-free WBT analysis framework, TractoFormer, that leverages tractography information at the level of individual fiber streamlines and provides a natural mechanism for interpretation of results using the attention mechanism of transformers. TractoFormer includes two main contributions. First, we propose a novel and simple 2D image representation of WBT, TractoEmbedding, to encode 3D fiber spatial relationships and any feature of interest that can be computed from individual fibers (such as FA or MD). Second, we design a network based on vision transformers (ViTs) that includes: 1) data augmentation to overcome model overfitting on small datasets, 2) identification of discriminative fibers for interpretation of results, and 3) ensemble learning to leverage fiber information from different brain regions. In a synthetic data experiment, TractoFormer successfully identifies discriminative fibers with simulated group differences. In a disease classification experiment comparing several methods, TractoFormer achieves the highest accuracy in classifying schizophrenia vs control. Discriminative fibers are identified in left hemispheric frontal and parietal superficial white matter regions, which have previously been shown to be affected in schizophrenia patients.

A Tutorial on the Spectral Theory of Markov Chains

Markov chains are a class of probabilistic models that have achieved widespread application in the quantitative sciences. This is in part due to their versatility, but is compounded by the ease with which they can be probed analytically. This tutorial provides an in-depth introduction to Markov chains, and explores their connection to graphs and random walks. We utilize tools from linear algebra and graph theory to describe the transition matrices of different types of Markov chains, with a particular focus on exploring properties of the eigenvalues and eigenvectors corresponding to these matrices. The results presented are relevant to a number of methods in machine learning and data mining, which we describe at various stages. Rather than being a novel academic study in its own right, this text presents a collection of known results, together with some new concepts. Moreover, the tutorial focuses on offering intuition to readers rather than formal understanding, and only assumes basic exposure to concepts from linear algebra and probability theory. It is therefore accessible to students and researchers from a wide variety of disciplines.

Implementing Reinforcement Learning Datacenter Congestion Control in NVIDIA NICs

Cloud datacenters are exponentially growing both in numbers and size. This increase results in a network activity surge that warrants better congestion avoidance. The resulting challenge is two-fold: (i) designing algorithms that can be custom-tuned to the complex traffic patterns of a given datacenter; but, at the same time (ii) run on low-level hardware with the required low latency of effective Congestion Control (CC). In this work, we present a Reinforcement Learning (RL) based CC solution that learns from certain traffic scenarios and successfully generalizes to others. We then distill the RL neural network policy into binary decision trees to achieve the desired $\mu$sec decision latency required for real-time inference with RDMA. We deploy the distilled policy on NVIDIA NICs in a real network and demonstrate state-of-the-art performance, balancing all tested metrics simultaneously: bandwidth, latency, fairness, and packet drops.

Cooperative Distribution Alignment via JSD Upper Bound

Unsupervised distribution alignment estimates a transformation that maps two or more source distributions to a shared aligned distribution given only samples from each distribution. This task has many applications including generative modeling, unsupervised domain adaptation, and socially aware learning. Most prior works use adversarial learning (i.e., min-max optimization), which can be challenging to optimize and evaluate. A few recent works explore non-adversarial flow-based (i.e., invertible) approaches, but they lack a unified perspective and are limited in efficiently aligning multiple distributions. Therefore, we propose to unify and generalize previous flow-based approaches under a single non-adversarial framework, which we prove is equivalent to minimizing an upper bound on the Jensen-Shannon Divergence (JSD). Importantly, our problem reduces to a min-min, i.e., cooperative, problem and can provide a natural evaluation metric for unsupervised distribution alignment. We present empirical results of our framework on both simulated and real-world datasets to demonstrate the benefits of our approach.

Towards Realistic Semi-Supervised Learning

Deep learning is pushing the state-of-the-art in many computer vision applications. However, it relies on large annotated data repositories, and capturing the unconstrained nature of the real-world data is yet to be solved. Semi-supervised learning (SSL) complements the annotated training data with a large corpus of unlabeled data to reduce annotation cost. The standard SSL approach assumes unlabeled data are from the same distribution as annotated data. Recently, ORCA [9] introduce a more realistic SSL problem, called open-world SSL, by assuming that the unannotated data might contain samples from unknown classes. This work proposes a novel approach to tackle SSL in open-world setting, where we simultaneously learn to classify known and unknown classes. At the core of our method, we utilize sample uncertainty and incorporate prior knowledge about class distribution to generate reliable pseudo-labels for unlabeled data belonging to both known and unknown classes. Our extensive experimentation showcases the effectiveness of our approach on several benchmark datasets, where it substantially outperforms the existing state-of-the-art on seven diverse datasets including CIFAR-100 (17.6%), ImageNet-100 (5.7%), and Tiny ImageNet (9.9%).

Ultra-Low-Bitrate Speech Coding with Pretrained Transformers

Speech coding facilitates the transmission of speech over low-bandwidth networks with minimal distortion. Neural-network based speech codecs have recently demonstrated significant improvements in quality over traditional approaches. While this new generation of codecs is capable of synthesizing high-fidelity speech, their use of recurrent or convolutional layers often restricts their effective receptive fields, which prevents them from compressing speech efficiently. We propose to further reduce the bitrate of neural speech codecs through the use of pretrained Transformers, capable of exploiting long-range dependencies in the input signal due to their inductive bias. As such, we use a pretrained Transformer in tandem with a convolutional encoder, which is trained end-to-end with a quantizer and a generative adversarial net decoder. Our numerical experiments show that supplementing the convolutional encoder of a neural speech codec with Transformer speech embeddings yields a speech codec with a bitrate of $600\,\mathrm{bps}$ that outperforms the original neural speech codec in synthesized speech quality when trained at the same bitrate. Subjective human evaluations suggest that the quality of the resulting codec is comparable or better than that of conventional codecs operating at three to four times the rate.

OpenLDN: Learning to Discover Novel Classes for Open-World Semi-Supervised Learning

Semi-supervised learning (SSL) is one of the dominant approaches to address the annotation bottleneck of supervised learning. Recent SSL methods can effectively leverage a large repository of unlabeled data to improve performance while relying on a small set of labeled data. One common assumption in most SSL methods is that the labeled and unlabeled data are from the same underlying data distribution. However, this is hardly the case in many real-world scenarios, which limits their applicability. In this work, instead, we attempt to solve the recently proposed challenging open-world SSL problem that does not make such an assumption. In the open-world SSL problem, the objective is to recognize samples of known classes, and simultaneously detect and cluster samples belonging to novel classes present in unlabeled data. This work introduces OpenLDN that utilizes a pairwise similarity loss to discover novel classes. Using a bi-level optimization rule this pairwise similarity loss exploits the information available in the labeled set to implicitly cluster novel class samples, while simultaneously recognizing samples from known classes. After discovering novel classes, OpenLDN transforms the open-world SSL problem into a standard SSL problem to achieve additional performance gains using existing SSL methods. Our extensive experiments demonstrate that OpenLDN outperforms the current state-of-the-art methods on multiple popular classification benchmarks while providing a better accuracy/training time trade-off.

Putting the Con in Context: Identifying Deceptive Actors in the Game of Mafia

While neural networks demonstrate a remarkable ability to model linguistic content, capturing contextual information related to a speaker's conversational role is an open area of research. In this work, we analyze the effect of speaker role on language use through the game of Mafia, in which participants are assigned either an honest or a deceptive role. In addition to building a framework to collect a dataset of Mafia game records, we demonstrate that there are differences in the language produced by players with different roles. We confirm that classification models are able to rank deceptive players as more suspicious than honest ones based only on their use of language. Furthermore, we show that training models on two auxiliary tasks outperforms a standard BERT-based text classification approach. We also present methods for using our trained models to identify features that distinguish between player roles, which could be used to assist players during the Mafia game.

Learning Task Embeddings for Teamwork Adaptation in Multi-Agent Reinforcement Learning

Successful deployment of multi-agent reinforcement learning often requires agents to adapt their behaviour. In this work, we discuss the problem of teamwork adaptation in which a team of agents needs to adapt their policies to solve novel tasks with limited fine-tuning. Motivated by the intuition that agents need to be able to identify and distinguish tasks in order to adapt their behaviour to the current task, we propose to learn multi-agent task embeddings (MATE). These task embeddings are trained using an encoder-decoder architecture optimised for reconstruction of the transition and reward functions which uniquely identify tasks. We show that a team of agents is able to adapt to novel tasks when provided with task embeddings. We propose three MATE training paradigms: independent MATE, centralised MATE, and mixed MATE which vary in the information used for the task encoding. We show that the embeddings learned by MATE identify tasks and provide useful information which agents leverage during adaptation to novel tasks.

State-Augmented Learnable Algorithms for Resource Management in Wireless Networks

We consider resource management problems in multi-user wireless networks, which can be cast as optimizing a network-wide utility function, subject to constraints on the long-term average performance of users across the network. We propose a state-augmented algorithm for solving the aforementioned radio resource management (RRM) problems, where, alongside the instantaneous network state, the RRM policy takes as input the set of dual variables corresponding to the constraints, which evolve depending on how much the constraints are violated during execution. We theoretically show that the proposed state-augmented algorithm leads to feasible and near-optimal RRM decisions. Moreover, focusing on the problem of wireless power control using graph neural network (GNN) parameterizations, we demonstrate the superiority of the proposed RRM algorithm over baseline methods across a suite of numerical experiments.

Guiding Machine Perception with Psychophysics

{G}{ustav} Fechner's 1860 delineation of psychophysics, the measurement of sensation in relation to its stimulus, is widely considered to be the advent of modern psychological science. In psychophysics, a researcher parametrically varies some aspects of a stimulus, and measures the resulting changes in a human subject's experience of that stimulus; doing so gives insight to the determining relationship between a sensation and the physical input that evoked it. This approach is used heavily in perceptual domains, including signal detection, threshold measurement, and ideal observer analysis. Scientific fields like vision science have always leaned heavily on the methods and procedures of psychophysics, but there is now growing appreciation of them by machine learning researchers, sparked by widening overlap between biological and artificial perception \cite{rojas2011automatic, scheirer2014perceptual,escalera2014chalearn,zhang2018agil, grieggs2021measuring}. Machine perception that is guided by behavioral measurements, as opposed to guidance restricted to arbitrarily assigned human labels, has significant potential to fuel further progress in artificial intelligence.

Improving Trustworthiness of AI Disease Severity Rating in Medical Imaging with Ordinal Conformal Prediction Sets

The regulatory approval and broad clinical deployment of medical AI have been hampered by the perception that deep learning models fail in unpredictable and possibly catastrophic ways. A lack of statistically rigorous uncertainty quantification is a significant factor undermining trust in AI results. Recent developments in distribution-free uncertainty quantification present practical solutions for these issues by providing reliability guarantees for black-box models on arbitrary data distributions as formally valid finite-sample prediction intervals. Our work applies these new uncertainty quantification methods -- specifically conformal prediction -- to a deep-learning model for grading the severity of spinal stenosis in lumbar spine MRI. We demonstrate a technique for forming ordinal prediction sets that are guaranteed to contain the correct stenosis severity within a user-defined probability (confidence interval). On a dataset of 409 MRI exams processed by the deep-learning model, the conformal method provides tight coverage with small prediction set sizes. Furthermore, we explore the potential clinical applicability of flagging cases with high uncertainty predictions (large prediction sets) by quantifying an increase in the prevalence of significant imaging abnormalities (e.g. motion artifacts, metallic artifacts, and tumors) that could degrade confidence in predictive performance when compared to a random sample of cases.

Offline RL Policies Should be Trained to be Adaptive

Offline RL algorithms must account for the fact that the dataset they are provided may leave many facets of the environment unknown. The most common way to approach this challenge is to employ pessimistic or conservative methods, which avoid behaviors that are too dissimilar from those in the training dataset. However, relying exclusively on conservatism has drawbacks: performance is sensitive to the exact degree of conservatism, and conservative objectives can recover highly suboptimal policies. In this work, we propose that offline RL methods should instead be adaptive in the presence of uncertainty. We show that acting optimally in offline RL in a Bayesian sense involves solving an implicit POMDP. As a result, optimal policies for offline RL must be adaptive, depending not just on the current state but rather all the transitions seen so far during evaluation.We present a model-free algorithm for approximating this optimal adaptive policy, and demonstrate the efficacy of learning such adaptive policies in offline RL benchmarks.

Data-driven synchronization-avoiding algorithms in the explicit distributed structural analysis of soft tissue

We propose a data-driven framework to increase the computational efficiency of the explicit finite element method in the structural analysis of soft tissue. An encoder-decoder long short-term memory deep neural network is trained based on the data produced by an explicit, distributed finite element solver. We leverage this network to predict synchronized displacements at shared nodes, minimizing the amount of communication between processors. We perform extensive numerical experiments to quantify the accuracy and stability of the proposed synchronization-avoiding algorithm.

CEN : Cooperatively Evolving Networks

A finitely repeated game is a dynamic game in which a simultaneous game is played finitely many times. GANs contain two competing modules: the generator module is trained to generate new examples, and the discriminator module is trained to discriminate real examples from generated examples. Training procedure of GAN is a finitely repeated game in which each module tries to optimize it's error at every instance of simultaneous game in a non-cooperative manner. We observed that we can achieve more accurate training, if at each instance of simultaneous game the stronger module cooperate with weaker module and only weaker module only optimize it's error.

Accelerating Hamiltonian Monte Carlo via Chebyshev Integration Time

Hamiltonian Monte Carlo (HMC) is a popular method in sampling. While there are quite a few works of studying this method on various aspects, an interesting question is how to choose its integration time to achieve acceleration. In this work, we consider accelerating the process of sampling from a distribution $\pi(x) \propto \exp(-f(x))$ via HMC via time-varying integration time. When the potential $f$ is $L$-smooth and $m$-strongly convex, i.e.\ for sampling from a log-smooth and strongly log-concave target distribution $\pi$, it is known that under a constant integration time, the number of iterations that ideal HMC takes to get an $\epsilon$ Wasserstein-2 distance to the target $\pi$ is $O( \kappa \log \frac{1}{\epsilon} )$, where $\kappa := \frac{L}{m}$ is the condition number. We propose a scheme of time-varying integration time based on the roots of Chebyshev polynomials. We show that in the case of quadratic potential $f$, i.e., when the target $\pi$ is a Gaussian distribution, ideal HMC with this choice of integration time only takes $O( \sqrt{\kappa} \log \frac{1}{\epsilon} )$ number of iterations to reach Wasserstein-2 distance less than $\epsilon$; this improvement on the dependence on condition number is akin to acceleration in optimization. The design and analysis of HMC with the proposed integration time is built on the tools of Chebyshev polynomials. Experiments find the advantage of adopting our scheme of time-varying integration time even for sampling from distributions with smooth strongly convex potentials that are not quadratic.

NeuralPassthrough: Learned Real-Time View Synthesis for VR

Virtual reality (VR) headsets provide an immersive, stereoscopic visual experience, but at the cost of blocking users from directly observing their physical environment. Passthrough techniques are intended to address this limitation by leveraging outward-facing cameras to reconstruct the images that would otherwise be seen by the user without the headset. This is inherently a real-time view synthesis challenge, since passthrough cameras cannot be physically co-located with the eyes. Existing passthrough techniques suffer from distracting reconstruction artifacts, largely due to the lack of accurate depth information (especially for near-field and disoccluded objects), and also exhibit limited image quality (e.g., being low resolution and monochromatic). In this paper, we propose the first learned passthrough method and assess its performance using a custom VR headset that contains a stereo pair of RGB cameras. Through both simulations and experiments, we demonstrate that our learned passthrough method delivers superior image quality compared to state-of-the-art methods, while meeting strict VR requirements for real-time, perspective-correct stereoscopic view synthesis over a wide field of view for desktop-connected headsets.

CLEAR: Improving Vision-Language Navigation with Cross-Lingual, Environment-Agnostic Representations

Vision-and-Language Navigation (VLN) tasks require an agent to navigate through the environment based on language instructions. In this paper, we aim to solve two key challenges in this task: utilizing multilingual instructions for improved instruction-path grounding and navigating through new environments that are unseen during training. To address these challenges, we propose CLEAR: Cross-Lingual and Environment-Agnostic Representations. First, our agent learns a shared and visually-aligned cross-lingual language representation for the three languages (English, Hindi and Telugu) in the Room-Across-Room dataset. Our language representation learning is guided by text pairs that are aligned by visual information. Second, our agent learns an environment-agnostic visual representation by maximizing the similarity between semantically-aligned image pairs (with constraints on object-matching) from different environments. Our environment agnostic visual representation can mitigate the environment bias induced by low-level visual information. Empirically, on the Room-Across-Room dataset, we show that our multilingual agent gets large improvements in all metrics over the strong baseline model when generalizing to unseen environments with the cross-lingual language representation and the environment-agnostic visual representation. Furthermore, we show that our learned language and visual representations can be successfully transferred to the Room-to-Room and Cooperative Vision-and-Dialogue Navigation task, and present detailed qualitative and quantitative generalization and grounding analysis. Our code is available at https://github.com/jialuli-luka/CLEAR

An Approximation Method for Fitted Random Forests

Random Forests (RF) is a popular machine learning method for classification and regression problems. It involves a bagging application to decision tree models. One of the primary advantages of the Random Forests model is the reduction in the variance of the forecast. In large scale applications of the model with millions of data points and hundreds of features, the size of the fitted objects can get very large and reach the limits on the available space in production setups, depending on the number and depth of the trees. This could be especially challenging when trained models need to be downloaded on-demand to small devices with limited memory. There is a need to approximate the trained RF models to significantly reduce the model size without losing too much of prediction accuracy. In this project we study methods that approximate each fitted tree in the Random Forests model using the multinomial allocation of the data points to the leafs. Specifically, we begin by studying whether fitting a multinomial logistic regression (and subsequently, a generalized additive model (GAM) extension) to the output of each tree helps reduce the size while preserving the prediction quality.

ST-CoNAL: Consistency-Based Acquisition Criterion Using Temporal Self-Ensemble for Active Learning

Modern deep learning has achieved great success in various fields. However, it requires the labeling of huge amounts of data, which is expensive and labor-intensive. Active learning (AL), which identifies the most informative samples to be labeled, is becoming increasingly important to maximize the efficiency of the training process. The existing AL methods mostly use only a single final fixed model for acquiring the samples to be labeled. This strategy may not be good enough in that the structural uncertainty of a model for given training data is not considered to acquire the samples. In this study, we propose a novel acquisition criterion based on temporal self-ensemble generated by conventional stochastic gradient descent (SGD) optimization. These self-ensemble models are obtained by capturing the intermediate network weights obtained through SGD iterations. Our acquisition function relies on a consistency measure between the student and teacher models. The student models are given a fixed number of temporal self-ensemble models, and the teacher model is constructed by averaging the weights of the student models. Using the proposed acquisition criterion, we present an AL algorithm, namely student-teacher consistency-based AL (ST-CoNAL). Experiments conducted for image classification tasks on CIFAR-10, CIFAR-100, Caltech-256, and Tiny ImageNet datasets demonstrate that the proposed ST-CoNAL achieves significantly better performance than the existing acquisition methods. Furthermore, extensive experiments show the robustness and effectiveness of our methods.

DBN-Mix: Training Dual Branch Network Using Bilateral Mixup Augmentation for Long-Tailed Visual Recognition

There is a growing interest in the challenging visual perception task of learning from long-tailed class distributions. The extreme class imbalance in the training dataset biases the model to prefer to recognize majority-class data over minority-class data. Recently, the dual branch network (DBN) framework has been proposed, where two branch networks; the conventional branch and the re-balancing branch were employed to improve the accuracy of long-tailed visual recognition. The re-balancing branch uses a reverse sampler to generate class-balanced training samples to mitigate bias due to class imbalance. Although this strategy has been quite successful in handling bias, using a reversed sampler for training can degrade the representation learning performance. To alleviate this issue, the conventional method used a carefully designed cumulative learning strategy, in which the influence of the re-balancing branch gradually increases throughout the entire training phase. In this study, we aim to develop a simple yet effective method to improve the performance of DBN without cumulative learning that is difficult to optimize. We devise a simple data augmentation method termed bilateral mixup augmentation, which combines one sample from the uniform sampler with another sample from the reversed sampler to produce a training sample. Furthermore, we present class-conditional temperature scaling that mitigates bias toward the majority class for the proposed DBN architecture. Our experiments performed on widely used long-tailed visual recognition datasets show that bilateral mixup augmentation is quite effective in improving the representation learning performance of DBNs, and that the proposed method achieves state-of-the-art performance for some categories.

Probability density estimation for sets of large graphs with respect to spectral information using stochastic block models

For graph-valued data sampled iid from a distribution $\mu$, the sample moments are computed with respect to a choice of metric. In this work, we equip the set of graphs with the pseudo-metric defined by the $\ell_2$ norm between the eigenvalues of the respective adjacency matrices. We use this pseudo metric and the respective sample moments of a graph valued data set to infer the parameters of a distribution $\hat{\mu}$ and interpret this distribution as an approximation of $\mu$. We verify experimentally that complex distributions $\mu$ can be approximated well taking this approach.

Automatic inspection of cultural monuments using deep and tensor-based learning on hyperspectral imagery

In Cultural Heritage, hyperspectral images are commonly used since they provide extended information regarding the optical properties of materials. Thus, the processing of such high-dimensional data becomes challenging from the perspective of machine learning techniques to be applied. In this paper, we propose a Rank-$R$ tensor-based learning model to identify and classify material defects on Cultural Heritage monuments. In contrast to conventional deep learning approaches, the proposed high order tensor-based learning demonstrates greater accuracy and robustness against overfitting. Experimental results on real-world data from UNESCO protected areas indicate the superiority of the proposed scheme compared to conventional deep learning models.

UniCR: Universally Approximated Certified Robustness via Randomized Smoothing

We study certified robustness of machine learning classifiers against adversarial perturbations. In particular, we propose the first universally approximated certified robustness (UniCR) framework, which can approximate the robustness certification of any input on any classifier against any $\ell_p$ perturbations with noise generated by any continuous probability distribution. Compared with the state-of-the-art certified defenses, UniCR provides many significant benefits: (1) the first universal robustness certification framework for the above 4 'any's; (2) automatic robustness certification that avoids case-by-case analysis, (3) tightness validation of certified robustness, and (4) optimality validation of noise distributions used by randomized smoothing. We conduct extensive experiments to validate the above benefits of UniCR and the advantages of UniCR over state-of-the-art certified defenses against $\ell_p$ perturbations.

Swin Deformable Attention U-Net Transformer (SDAUT) for Explainable Fast MRI

Fast MRI aims to reconstruct a high fidelity image from partially observed measurements. Exuberant development in fast MRI using deep learning has been witnessed recently. Meanwhile, novel deep learning paradigms, e.g., Transformer based models, are fast-growing in natural language processing and promptly developed for computer vision and medical image analysis due to their prominent performance. Nevertheless, due to the complexity of the Transformer, the application of fast MRI may not be straightforward. The main obstacle is the computational cost of the self-attention layer, which is the core part of the Transformer, can be expensive for high resolution MRI inputs. In this study, we propose a new Transformer architecture for solving fast MRI that coupled Shifted Windows Transformer with U-Net to reduce the network complexity. We incorporate deformable attention to construe the explainability of our reconstruction model. We empirically demonstrate that our method achieves consistently superior performance on the fast MRI task. Besides, compared to state-of-the-art Transformer models, our method has fewer network parameters while revealing explainability. The code is publicly available at https://github.com/ayanglab/SDAUT.

Deterministic Decoupling of Global Features and its Application to Data Analysis

We introduce a method for deterministic decoupling of global features and show its applicability to improve data analysis performance, as well as to open new venues for feature transfer. We propose a new formalism that is based on defining transformations on submanifolds, by following trajectories along the features gradients. Through these transformations we define a normalization that, we demonstrate, allows for decoupling differentiable features. By applying this to sampling moments, we obtain a quasi-analytic solution for the orthokurtosis, a normalized version of the kurtosis that is not just decoupled from mean and variance, but also from skewness. We apply this method in the original data domain and at the output of a filter bank to regression and classification problems based on global descriptors, obtaining a consistent and significant improvement in performance as compared to using classical (non-decoupled) descriptors.

A survey of multimodal deep generative models

Multimodal learning is a framework for building models that make predictions based on different types of modalities. Important challenges in multimodal learning are the inference of shared representations from arbitrary modalities and cross-modal generation via these representations; however, achieving this requires taking the heterogeneous nature of multimodal data into account. In recent years, deep generative models, i.e., generative models in which distributions are parameterized by deep neural networks, have attracted much attention, especially variational autoencoders, which are suitable for accomplishing the above challenges because they can consider heterogeneity and infer good representations of data. Therefore, various multimodal generative models based on variational autoencoders, called multimodal deep generative models, have been proposed in recent years. In this paper, we provide a categorized survey of studies on multimodal deep generative models.

Adapting to Online Label Shift with Provable Guarantees

The standard supervised learning paradigm works effectively when training data shares the same distribution as the upcoming testing samples. However, this assumption is often violated in real-world applications, especially when testing data appear in an online fashion. In this paper, we formulate and investigate the problem of online label shift (OLaS): the learner trains an initial model from the labeled offline data and then deploys it to an unlabeled online environment where the underlying label distribution changes over time but the label-conditional density does not. The non-stationarity nature and the lack of supervision make the problem challenging to be tackled. To address the difficulty, we construct a new unbiased risk estimator that utilizes the unlabeled data, which exhibits many benign properties albeit with potential non-convexity. Building upon that, we propose novel online ensemble algorithms to deal with the non-stationarity of the environments. Our approach enjoys optimal dynamic regret, indicating that the performance is competitive with a clairvoyant who knows the online environments in hindsight and then chooses the best decision for each round. The obtained dynamic regret bound scales with the intensity and pattern of label distribution shift, hence exhibiting the adaptivity in the OLaS problem. Extensive experiments are conducted to validate the effectiveness and support our theoretical findings.

Improving Covariance Conditioning of the SVD Meta-layer by Orthogonality

Inserting an SVD meta-layer into neural networks is prone to make the covariance ill-conditioned, which could harm the model in the training stability and generalization abilities. In this paper, we systematically study how to improve the covariance conditioning by enforcing orthogonality to the Pre-SVD layer. Existing orthogonal treatments on the weights are first investigated. However, these techniques can improve the conditioning but would hurt the performance. To avoid such a side effect, we propose the Nearest Orthogonal Gradient (NOG) and Optimal Learning Rate (OLR). The effectiveness of our methods is validated in two applications: decorrelated Batch Normalization (BN) and Global Covariance Pooling (GCP). Extensive experiments on visual recognition demonstrate that our methods can simultaneously improve the covariance conditioning and generalization. Moreover, the combinations with orthogonal weight can further boost the performances.

An Intrusion Detection System based on Deep Belief Networks

The rapid growth of connected devices has led to the proliferation of novel cyber-security threats known as zero-day attacks. Traditional behaviour-based IDS rely on DNN to detect these attacks. The quality of the dataset used to train the DNN plays a critical role in the detection performance, with underrepresented samples causing poor performances. In this paper, we develop and evaluate the performance of DBN on detecting cyber-attacks within a network of connected devices. The CICIDS2017 dataset was used to train and evaluate the performance of our proposed DBN approach. Several class balancing techniques were applied and evaluated. Lastly, we compare our approach against a conventional MLP model and the existing state-of-the-art. Our proposed DBN approach shows competitive and promising results, with significant performance improvement on the detection of attacks underrepresented in the training dataset.

An Empirical Study of Implicit Regularization in Deep Offline RL

Deep neural networks are the most commonly used function approximators in offline Reinforcement Learning these days. Prior works have shown that neural nets trained with TD-learning and gradient descent can exhibit implicit regularization that can be characterized by under-parameterization of these networks. Specifically, the rank of the penultimate feature layer, also called \textit{effective rank}, has been observed to drastically collapse during the training. In turn, this collapse has been argued to reduce the model's ability to further adapt in later stages of learning, leading to the diminished final performance. Such an association between the effective rank and performance makes effective rank compelling for offline RL, primarily for offline policy evaluation. In this work, we conduct a careful empirical study on the relation between effective rank and performance on three offline RL datasets : bsuite, Atari, and DeepMind lab. We observe that a direct association exists only in restricted settings and disappears in the more extensive hyperparameter sweeps. Also, we empirically identify three phases of learning that explain the impact of implicit regularization on the learning dynamics and found that bootstrapping alone is insufficient to explain the collapse of the effective rank. Further, we show that several other factors could confound the relationship between effective rank and performance and conclude that studying this association under simplistic assumptions could be highly misleading.

Neural Networks and the Chomsky Hierarchy

Reliable generalization lies at the heart of safe ML and AI. However, understanding when and how neural networks generalize remains one of the most important unsolved problems in the field. In this work, we conduct an extensive empirical study (2200 models, 16 tasks) to investigate whether insights from the theory of computation can predict the limits of neural network generalization in practice. We demonstrate that grouping tasks according to the Chomsky hierarchy allows us to forecast whether certain architectures will be able to generalize to out-of-distribution inputs. This includes negative results where even extensive amounts of data and training time never led to any non-trivial generalization, despite models having sufficient capacity to perfectly fit the training data. Our results show that, for our subset of tasks, RNNs and Transformers fail to generalize on non-regular tasks, LSTMs can solve regular and counter-language tasks, and only networks augmented with structured memory (such as a stack or memory tape) can successfully generalize on context-free and context-sensitive tasks.

Predicting Out-of-Domain Generalization with Local Manifold Smoothness

Understanding how machine learning models generalize to new environments is a critical part of their safe deployment. Recent work has proposed a variety of complexity measures that directly predict or theoretically bound the generalization capacity of a model. However, these methods rely on a strong set of assumptions that in practice are not always satisfied. Motivated by the limited settings in which existing measures can be applied, we propose a novel complexity measure based on the local manifold smoothness of a classifier. We define local manifold smoothness as a classifier's output sensitivity to perturbations in the manifold neighborhood around a given test point. Intuitively, a classifier that is less sensitive to these perturbations should generalize better. To estimate smoothness we sample points using data augmentation and measure the fraction of these points classified into the majority class. Our method only requires selecting a data augmentation method and makes no other assumptions about the model or data distributions, meaning it can be applied even in out-of-domain (OOD) settings where existing methods cannot. In experiments on robustness benchmarks in image classification, sentiment analysis, and natural language inference, we demonstrate a strong and robust correlation between our manifold smoothness measure and actual OOD generalization on over 3,000 models evaluated on over 100 train/test domain pairs.

Learning to Accelerate Approximate Methods for Solving Integer Programming via Early Fixing

Integer programming (IP) is an important and challenging problem. Approximate methods have shown promising performance on both effectiveness and efficiency for solving the IP problem. However, we observed that a large fraction of variables solved by some iterative approximate methods fluctuate around their final converged discrete states in very long iterations. Inspired by this observation, we aim to accelerate these approximate methods by early fixing these fluctuated variables to their converged states while not significantly harming the solution accuracy. To this end, we propose an early fixing framework along with the approximate method. We formulate the whole early fixing process as a Markov decision process, and train it using imitation learning. A policy network will evaluate the posterior probability of each free variable concerning its discrete candidate states in each block of iterations. Specifically, we adopt the powerful multi-headed attention mechanism in the policy network. Extensive experiments on our proposed early fixing framework are conducted to three different IP applications: constrained linear programming, MRF energy minimization and sparse adversarial attack. The former one is linear IP problem, while the latter two are quadratic IP problems. We extend the problem scale from regular size to significantly large size. The extensive experiments reveal the competitiveness of our early fixing framework: the runtime speeds up significantly, while the solution quality does not degrade much, even in some cases it is available to obtain better solutions. Our proposed early fixing framework can be regarded as an acceleration extension of ADMM methods for solving integer programming. The source codes are available at \url{https://github.com/SCLBD/Accelerated-Lpbox-ADMM}.

Resource Allocation in Multicore Elastic Optical Networks: A Deep Reinforcement Learning Approach

A deep reinforcement learning approach is applied, for the first time, to solve the routing, modulation, spectrum and core allocation (RMSCA) problem in dynamic multicore fiber elastic optical networks (MCF-EONs). To do so, a new environment - compatible with OpenAI's Gym - was designed and implemented to emulate the operation of MCF-EONs. The new environment processes the agent actions (selection of route, core and spectrum slot) by considering the network state and physical-layer-related aspects. The latter includes the available modulation formats and their reach and the inter-core crosstalk (XT), an MCF-related impairment. If the resulting quality of the signal is acceptable, the environment allocates the resources selected by the agent. After processing the agent's action, the environment is configured to give the agent a numerical reward and information about the new network state. The blocking performance of four different agents was compared through simulation to 3 baseline heuristics used in MCF-EONs. Results obtained for the NSFNet and COST239 network topologies show that the best-performing agent achieves, on average, up to a four-times decrease in blocking probability concerning the best-performing baseline heuristic methods.

Image Amodal Completion: A Survey

Existing computer vision systems can compete with humans in understanding the visible parts of objects, but still fall far short of humans when it comes to depicting the invisible parts of partially occluded objects. Image amodal completion aims to equip computers with human-like amodal completion functions to understand an intact object despite it being partially occluded. The main purpose of this survey is to provide an intuitive understanding of the research hotspots, key technologies and future trends in the field of image amodal completion. Firstly, we present a comprehensive review of the latest literature in this emerging field, exploring three key tasks in image amodal completion, including amodal shape completion, amodal appearance completion, and order perception. Then we examine popular datasets related to image amodal completion along with their common data collection methods and evaluation metrics. Finally, we discuss real-world applications and future research directions for image amodal completion, facilitating the reader's understanding of the challenges of existing technologies and upcoming research trends.

PRoA: A Probabilistic Robustness Assessment against Functional Perturbations

In safety-critical deep learning applications robustness measurement is a vital pre-deployment phase. However, existing robustness verification methods are not sufficiently practical for deploying machine learning systems in the real world. On the one hand, these methods attempt to claim that no perturbations can ``fool'' deep neural networks (DNNs), which may be too stringent in practice. On the other hand, existing works rigorously consider $L_p$ bounded additive perturbations on the pixel space, although perturbations, such as colour shifting and geometric transformations, are more practically and frequently occurring in the real world. Thus, from the practical standpoint, we present a novel and general {\it probabilistic robustness assessment method} (PRoA) based on the adaptive concentration, and it can measure the robustness of deep learning models against functional perturbations. PRoA can provide statistical guarantees on the probabilistic robustness of a model, \textit{i.e.}, the probability of failure encountered by the trained model after deployment. Our experiments demonstrate the effectiveness and flexibility of PRoA in terms of evaluating the probabilistic robustness against a broad range of functional perturbations, and PRoA can scale well to various large-scale deep neural networks compared to existing state-of-the-art baselines. For the purpose of reproducibility, we release our tool on GitHub: \url{ https://github.com/TrustAI/PRoA}.

Robust Reinforcement Learning in Continuous Control Tasks with Uncertainty Set Regularization

Reinforcement learning (RL) is recognized as lacking generalization and robustness under environmental perturbations, which excessively restricts its application for real-world robotics. Prior work claimed that adding regularization to the value function is equivalent to learning a robust policy with uncertain transitions. Although the regularization-robustness transformation is appealing for its simplicity and efficiency, it is still lacking in continuous control tasks. In this paper, we propose a new regularizer named $\textbf{U}$ncertainty $\textbf{S}$et $\textbf{R}$egularizer (USR), by formulating the uncertainty set on the parameter space of the transition function. In particular, USR is flexible enough to be plugged into any existing RL framework. To deal with unknown uncertainty sets, we further propose a novel adversarial approach to generate them based on the value function. We evaluate USR on the Real-world Reinforcement Learning (RWRL) benchmark, demonstrating improvements in the robust performance for perturbed testing environments.

Towards trustworthy Energy Disaggregation: A review of challenges, methods and perspectives for Non-Intrusive Load Monitoring

Non-intrusive load monitoring (NILM) is the task of disaggregating the total power consumption into its individual sub-components. Over the years, signal processing and machine learning algorithms have been combined to achieve this. A lot of publications and extensive research works are performed on energy disaggregation or NILM for the state-of-the-art methods to reach on the desirable performance. The initial interest of the scientific community to formulate and describe mathematically the NILM problem using machine learning tools has now shifted into a more practical NILM. Nowadays, we are in the mature NILM period where there is an attempt for NILM to be applied in real-life application scenarios. Thus, complexity of the algorithms, transferability, reliability, practicality and in general trustworthiness are the main issues of interest. This review narrows the gap between the early immature NILM era and the mature one. In particular, the paper provides a comprehensive literature review of the NILM methods for residential appliances only. The paper analyzes, summarizes and presents the outcomes of a large number of recently published scholarly articles. Also, the paper discusses the highlights of these methods and introduces the research dilemmas that should be taken into consideration by researchers to apply NILM methods. Finally, we show the need for transferring the traditional disaggregation models into a practical and trustworthy framework.

The StarCraft Multi-Agent Challenges+ : Learning of Multi-Stage Tasks and Environmental Factors without Precise Reward Functions

In this paper, we propose a novel benchmark called the StarCraft Multi-Agent Challenges+, where agents learn to perform multi-stage tasks and to use environmental factors without precise reward functions. The previous challenges (SMAC) recognized as a standard benchmark of Multi-Agent Reinforcement Learning are mainly concerned with ensuring that all agents cooperatively eliminate approaching adversaries only through fine manipulation with obvious reward functions. This challenge, on the other hand, is interested in the exploration capability of MARL algorithms to efficiently learn implicit multi-stage tasks and environmental factors as well as micro-control. This study covers both offensive and defensive scenarios. In the offensive scenarios, agents must learn to first find opponents and then eliminate them. The defensive scenarios require agents to use topographic features. For example, agents need to position themselves behind protective structures to make it harder for enemies to attack. We investigate MARL algorithms under SMAC+ and observe that recent approaches work well in similar settings to the previous challenges, but misbehave in offensive scenarios. Additionally, we observe that an enhanced exploration approach has a positive effect on performance but is not able to completely solve all scenarios. This study proposes new directions for future research.

Disentangling private classes through regularization

Deep learning models are nowadays broadly deployed to solve an incredibly large variety of tasks. However, little attention has been devoted to connected legal aspects. In 2016, the European Union approved the General Data Protection Regulation which entered into force in 2018. Its main rationale was to protect the privacy and data protection of its citizens by the way of operating of the so-called "Data Economy". As data is the fuel of modern Artificial Intelligence, it is argued that the GDPR can be partly applicable to a series of algorithmic decision making tasks before a more structured AI Regulation enters into force. In the meantime, AI should not allow undesired information leakage deviating from the purpose for which is created. In this work we propose DisP, an approach for deep learning models disentangling the information related to some classes we desire to keep private, from the data processed by AI. In particular, DisP is a regularization strategy de-correlating the features belonging to the same private class at training time, hiding the information of private classes membership. Our experiments on state-of-the-art deep learning models show the effectiveness of DisP, minimizing the risk of extraction for the classes we desire to keep private.

Correlation between entropy and generalizability in a neural network

Although neural networks can solve very complex machine-learning problems, the theoretical reason for their generalizability is still not fully understood. Here we use Wang-Landau Mote Carlo algorithm to calculate the entropy (logarithm of the volume of a part of the parameter space) at a given test accuracy, and a given training loss function value or training accuracy. Our results show that entropical forces help generalizability. Although our study is on a very simple application of neural networks (a spiral dataset and a small, fully-connected neural network), our approach should be useful in explaining the generalizability of more complicated neural networks in future works.

Local Multi-Label Explanations for Random Forest

Multi-label classification is a challenging task, particularly in domains where the number of labels to be predicted is large. Deep neural networks are often effective at multi-label classification of images and textual data. When dealing with tabular data, however, conventional machine learning algorithms, such as tree ensembles, appear to outperform competition. Random forest, being a popular ensemble algorithm, has found use in a wide range of real-world problems. Such problems include fraud detection in the financial domain, crime hotspot detection in the legal sector, and in the biomedical field, disease probability prediction when patient records are accessible. Since they have an impact on people's lives, these domains usually require decision-making systems to be explainable. Random Forest falls short on this property, especially when a large number of tree predictors are used. This issue was addressed in a recent research named LionForests, regarding single label classification and regression. In this work, we adapt this technique to multi-label classification problems, by employing three different strategies regarding the labels that the explanation covers. Finally, we provide a set of qualitative and quantitative experiments to assess the efficacy of this approach.

Conflicting Interactions Among Protections Mechanisms for Machine Learning Models

Nowadays, systems based on machine learning (ML) are widely used in different domains. Given their popularity, ML models have become targets for various attacks. As a result, research at the intersection of security and privacy, and ML has flourished. The research community has been exploring the attack vectors and potential mitigations separately. However, practitioners will likely need to deploy defences against several threats simultaneously. A solution that is optimal for a specific concern may interact negatively with solutions intended to address other concerns. In this work, we explore the potential for conflicting interactions between different solutions that enhance the security/privacy of ML-base systems. We focus on model and data ownership; exploring how ownership verification techniques interact with other ML security/privacy techniques like differentially private training, and robustness against model evasion. We provide a framework, and conduct systematic analysis of pairwise interactions. We show that many pairs are incompatible. Where possible, we provide relaxations to the hyperparameters or the techniques themselves that allow for the simultaneous deployment. Lastly, we discuss the implications and provide guidelines for future work.

Unsupervised Crowdsourcing with Accuracy and Cost Guarantees

We consider the problem of cost-optimal utilization of a crowdsourcing platform for binary, unsupervised classification of a collection of items, given a prescribed error threshold. Workers on the crowdsourcing platform are assumed to be divided into multiple classes, based on their skill, experience, and/or past performance. We model each worker class via an unknown confusion matrix, and a (known) price to be paid per label prediction. For this setting, we propose algorithms for acquiring label predictions from workers, and for inferring the true labels of items. We prove that if the number of (unlabeled) items available is large enough, our algorithms satisfy the prescribed error thresholds, incurring a cost that is near-optimal. Finally, we validate our algorithms, and some heuristics inspired by them, through an extensive case study.