Reinforcement Learning on Atari-57 (full)
22HWRBGDI-H3
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| GDI-H3Training Scale=2.00E+08, Playtime=38.52022.06 | 22 | 9,620.33 | 1,146.39 | 154.27 | 50.63 | 71.26 | 50.63 | |
| MuZeroTraining Scale=2.00E+10, Playtime=38582022.06 | 19 | 4,994.97 | 2,041.12 | 152.1 | 49.8 | 71.94 | 49.8 | |
| Agent57Training Scale=1.00E+11, Playtime=192902022.06 | 18 | 4,762.17 | 1,933.49 | 125.92 | 43.62 | 76.26 | 43.62 | |
| GDI-I3Training Scale=2.00E+08, Playtime=38.52022.06 | 17 | 7,810.1 | 832.5 | 117.98 | 35.78 | 61.66 | 35.78 | |
| R2D2Training Scale=1.00E+10, Playtime=19292022.06 | 15 | 3,373.48 | 1,342.27 | 98.78 | 33.62 | 60.43 | 33.62 | |
| Go-ExploreTraining Scale=1.00E+10, Playtime=19292022.06 | 15 | 4,989.31 | 1,451.55 | 116.89 | 50.5 | 71.8 | 50.5 | |
| NGUTraining Scale=3.50E+10, Playtime=6751.52022.06 | 8 | 3,169.07 | 1,174.92 | 76 | 21.19 | 50.47 | 21.19 | |
| LASERTraining Scale=2.00E+08, Playtime=38.52022.06 | 7 | 1,740.94 | 454.91 | 45.39 | 8.08 | 36.78 | 8.08 | |
| MuesliTraining Scale=2.00E+08, Playtime=38.52022.06 | 5 | 2,538.12 | 1,077.47 | 75.52 | 24.86 | 48.74 | 24.86 | |
| RainbowTraining Scale=2.00E+08, Playtime=38.52022.06 | 4 | 873.54 | 230.99 | 28.39 | 4.92 | 28.39 | 4.92 | |
| IMPALATraining Scale=2.00E+08, Playtime=38.52022.06 | 3 | 956.99 | 191.82 | 34.52 | 4.31 | 29.45 | 4.31 | |
| Dreamer V2Training Scale=2.00E+08, Playtime=38.582022.06 | 3 | 642.49 | 178.04 | 38.6 | 4.29 | 27.73 | 4.29 | |
| SimPLeTraining Scale=1.00E+06, Playtime=0.192022.06 | 0 | 25.78 | 5.55 | 4.8 | 0.13 | 4.8 | 0.13 |