CoinRun on Procgen easy
9.2Mean Episode ReturnTRPO-KL
Evaluation Results
| Method | Links | |
|---|---|---|
| TRPO-KLTraining steps=25M, Difficulty mode=easy, Seeds=3, Eval episodes per seed=252026.02 | 9.2 | |
| BPPOTraining steps=25M, Difficulty mode=easy, Seeds=3, Eval episodes per seed=252026.02 | 8.53 | |
| PPOTraining steps=25M, Difficulty mode=easy, Seeds=3, Eval episodes per seed=252026.02 | 8.13 | |
| BTRPO-BCTraining steps=25M, Difficulty mode=easy, Seeds=3, Eval episodes per seed=252026.02 | 4.87 |