Partially observable navigation on Minigrid Maze3
96Solved RateDR
Evaluation results
| Method | Links | |
|---|---|---|
| DRBase algorithm=LSTM-based PPO, Environmental steps=250M2024.10 | 96 | |
| PLRBase algorithm=LSTM-based PPO, Environmental steps=250M2024.10 | 90 | |
| ADD w/o guidanceBase algorithm=LSTM-based PPO, Environmental steps=250M, Guidance=false2024.10 | 77 | |
| ADDBase algorithm=LSTM-based PPO, Environmental steps=250M, Guidance=true2024.10 | 74 | |
| PAIREDBase algorithm=LSTM-based PPO, Environmental steps=250M2024.10 | 64 | |
| ACCELBase algorithm=LSTM-based PPO, Environmental steps=250M2024.10 | 61 |