Reinforcement Learning on MiniGrid v0 (test)
0.944GoToDoor-8x8 Success RateRDE+DQN
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| RDE+DQNRR=12023.10 | 0.944 | 0.215 | 0.159 | 0.159 | 0.793 | |
| RDE+DQNRR=22023.10 | 0.929 | 0.237 | 0.137 | 0.155 | 0.791 | |
| RDE+DQNRR=0.52023.10 | 0.911 | 0.248 | 0.186 | 0.148 | 0.793 | |
| SR+DQNRR=0.52023.10 | 0.71 | 0 | 0.012 | 0.03 | 0.655 | |
| DQNRR=0.52023.10 | 0.709 | 0.035 | 0 | 0.002 | 0.747 | |
| SR+DQNRR=22023.10 | 0.684 | 0 | 0.014 | 0.034 | 0.706 | |
| SR+DQNRR=12023.10 | 0.659 | 0.02 | 0.013 | 0.072 | 0.729 | |
| DQNRR=12023.10 | 0.544 | 0.016 | 0 | 0 | 0.761 | |
| DQNRR=22023.10 | 0.159 | 0.029 | 0 | 0 | 0.674 |