Reinforcement Learning on Atari 26 100K environment steps (Individual Scores & HNS)
7,127.7Alien ScoreHuman
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HumanEnvironment steps=100K2021.03 | 7,127.7 | 1,719.5 | 742 | 8,503.3 | 753.1 | 37,187.5 | 12.1 | 30.5 | 7,387.8 | 23,829.4 | 1,971 | 29.6 | 4,334.7 | 2,412.5 | 30,826.4 | 302.8 | 3,035 | 2,665.5 | 22,736.3 | 6,951.6 | 14.6 | 69,571.3 | 13,455 | 7,845 | 42,054.7 | 11,693.2 | 1 | 1 | — | |
| APTEnvironment steps=100K, Random seeds=102021.03 | 2,614.8 | 211.5 | 891.5 | 185.5 | 416.7 | 7,065.1 | 21.3 | 10.9 | 317 | 44,128 | 5,071.8 | 29.9 | 1,796.1 | 2,590.4 | 6,789.1 | 356.1 | 412 | 2,312 | 17,357 | 2,827.1 | -8 | 96.1 | 17,671.2 | 4,782.1 | 2,116.7 | 8,289.4 | 69.55 | 47.5 | 7 | |
| SPREnvironment steps=100K2021.03 | 801.5 | 176.3 | 571 | 977.8 | 380.9 | 16,651 | 35.8 | 17.1 | 974.8 | 42,923.6 | 1,283.4 | 24.4 | 1,821.5 | 715.2 | 7,019.2 | 365.4 | 3,276.4 | 2,688.9 | 13,192.7 | 1,313.2 | -5.9 | 124 | 669.1 | 14,220.5 | 583.1 | 28,138.5 | 70.4 | 41.5 | 7 | |
| DrQEnvironment steps=100K2021.03 | 771.2 | 102.8 | 452.4 | 603.5 | 168.9 | 12,954 | 6 | 16.1 | 780.3 | 20,516.5 | 443.9 | 9.8 | 331.1 | 636.3 | 3,736.3 | 236 | 940.6 | 4,018.1 | 9,111 | 960.5 | -8.5 | -13.6 | 854.4 | 8,895.1 | 301.2 | 3,180.8 | 35.7 | 26.8 | 2 | |
| DEREnvironment steps=100K2021.03 | 739.9 | 188.6 | 431.2 | 470.8 | 51 | 10,124.6 | 0.2 | 1.9 | 861.8 | 16,185.2 | 341.3 | 27.9 | 866.8 | 349.5 | 6,857 | 301.6 | 779.3 | 2,851.5 | 14,346.1 | 1,204.1 | -19.3 | 97.8 | 1,152.9 | 9,600 | 354.1 | 2,877.4 | 28.5 | 16.1 | 2 | |
| SimPLeEnvironment steps=100K2021.03 | 616.9 | 88 | 527.2 | 1,128.3 | 34.2 | 5,184.4 | 9.1 | 16.4 | 1,246.9 | 62,583.6 | 469.1 | 20.3 | 254.7 | 771 | 2,656.6 | 125.3 | 323.1 | 4,539.9 | 17,257.2 | 1,480 | 12.8 | 58.3 | 1,288.8 | 5,640.6 | 683.3 | 3,350.3 | 44.3 | 14.4 | 2 | |
| CURLEnvironment steps=100K2021.03 | 558.2 | 142.1 | 600.6 | 734.5 | 131.6 | 14,870 | 1.2 | 4.9 | 1,058.5 | 12,146.5 | 612.4 | 26.7 | 1,181.3 | 669.3 | 6,279.3 | 471 | 872.5 | 4,229.6 | 14,307.8 | 1,465.5 | -16.5 | 218.4 | 1,042.4 | 5,661 | 384.5 | 2,955.2 | 38.1 | 17.5 | 2 | |
| VISREnvironment steps=100K2021.03 | 364.4 | 186 | 12,091.1 | 6,216.7 | 71.3 | 7,072.7 | 13.4 | 17.9 | 800.8 | 49,373.9 | 8,994.9 | -12.1 | 230.9 | 498.6 | 663.5 | 484.4 | 1,761.9 | 3,142.5 | 16,754.9 | 558.5 | -26.2 | 98.3 | 666.3 | 6,146.7 | 706.6 | 10,037.6 | 64.31 | 12.36 | 6 | |
| RandomEnvironment steps=100K2021.03 | 227.8 | 5.8 | 222.4 | 210 | 14.2 | 2,360 | 0.1 | 1.7 | 811 | 10,780.5 | 152.1 | 0 | 65.2 | 257.6 | 1,027 | 29 | 52 | 1,598 | 258.5 | 307.3 | -20.7 | 24.9 | 163.9 | 11.5 | 68.4 | 533.4 | 0 | 0 | 0 |