Goal-conditioned Reinforcement Learning on Modified PandaReachDense Joint space, +N+G v3
-7.63Average RewardO2S
Evaluation results
| Method | Links | |
|---|---|---|
| O2SAction Space=Joint, Initial Joint Randomization=true, Goal Randomization=true2025.04 | -7.63 | |
| Q2SAction Space=Joint, Initial Joint Randomization=true, Goal Randomization=true2025.04 | -8.13 | |
| P2SAction Space=Joint, Initial Joint Randomization=true, Goal Randomization=true2025.04 | -8.17 | |
| RSAction Space=Joint, Initial Joint Randomization=true, Goal Randomization=true2025.04 | -16.13 |