Reinforcement Learning on PrivateEye
69,571Final Mean PerformanceAvg. Human
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Avg. Human2018.10 | 69,571 | — | |
| SOTAcitation=[2]2018.10 | 15,806 | — | |
| RND2018.10 | 8,666 | — | |
| PPO2018.10 | 105 | — | |
| Dynamics2018.10 | 33 | — | |
| Average HumanPure-exploration regime=true2022.06 | — | 69,571.3 | |
| BYOL-ExplorePure-exploration regime=true, Episodes=10, Seeds=32022.06 | — | 5,495.88 | |
| ICMPure-exploration regime=true, Episodes=10, Seeds=32022.06 | — | 469.09 | |
| RNDPure-exploration regime=true, Episodes=10, Seeds=32022.06 | — | 4,309.31 |