Continuous Control on Ant CrossMaze
1.76Goal DistanceDiPRL
Evaluation Results
| Method | Links | |
|---|---|---|
| DiPRL2026.05 | 1.76 | |
| πcont.-PRLmode=continuous2026.05 | 2.79 | |
| π-PRL2026.05 | 4.43 | |
| πdisc.-PRLmode=discrete2026.05 | 6.29 | |
| PPO2026.05 | 6.65 | |
| VIPER (PPO)2026.05 | 8.17 | |
| DTSemNets2026.05 | 8.37 |