YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
DQN on LunarLander-v3
This repository contains a Deep Q-Network (DQN) agent trained on LunarLander-v3 using PyTorch.
Architecture
- 2 hidden layers (128 units each)
- ReLU activations
- Target Network
- Experience Replay
- Gradient Clipping
Hyperparameters
- Learning rate: 1e-3
- Gamma: 0.99
- Batch size: 64
- Replay buffer size: 100000
- Target update frequency: 10 episodes
Performance
Average Test Reward: ~260
The environment is considered solved when reward > 200.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support