PPO Agent Playing LunarLander-v2
This is a PPO agent trained from scratch using PyTorch.
Environment
LunarLander-v2
Evaluation
Mean reward: -202.83 +/- 121.89
Training
Total timesteps: 10000
Learning rate: 0.00025
Gamma: 0.99
GAE lambda: 0.95
Clip coefficient: 0.2
Update epochs: 4
