REINFORCE Agent playing Pixelcopter-PLE-v0
This is a trained REINFORCE agent for Pixelcopter-PLE-v0.
Training
- Algorithm: REINFORCE
- Hidden size: 64
- Training episodes: 50000+
- Maximum steps per episode: 10000
- Gamma: 0.99
- Learning rate: 0.0001
- Fine-tuning learning rate: 0.00005
Evaluation
Mean reward: 40.27
Standard deviation: 31.68
Course score (mean - std): 8.60
Course requirement: >= 5
This model was trained for Unit 4 of the Hugging Face Deep Reinforcement Learning Course.
Evaluation results
- mean_reward on Pixelcopter-PLE-v0self-reported40.27 +/- 31.68