Reinforcement Learning on Continuous Atari learning environment (CALE) DonkeyKong
216,793Cumulative RewardH2RL++
Evaluation results
| Method | Links | |
|---|---|---|
| H2RL++2026.03 | 216,793 | |
| H2RL+2026.03 | 87,780 | |
| H2RL2026.03 | 33,690 | |
| H2RLAction Space=Continuous2026.03 | 10,818 | |
| PPO2026.03 | 4,536 | |
| PPOAction Space=Continuous2026.03 | 3,836 | |
| hPPOAction Space=Continuous2026.03 | 991 | |
| hReason2026.03 | 860 | |
| hReasonAction Space=Continuous2026.03 | 542 | |
| hDQN2026.03 | 434 | |
| hPPO2026.03 | 418 | |
| BlendRL2026.03 | 85 | |
| NUDGE2026.03 | 21 |