Offline Reinforcement Learning on Confounded Atari DQN Replay (test)
1,056.2Alien Score
OREO
Updated 1mo ago
Evaluation Results
| Method | Links | |||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2021.10 | 1,056.2 | 105.7 | 840.9 | 180.8 | 493.9 | 12,700 | 36.4 | 4.2 | 977.4 | 55,523.4 | 224.5 | 522.8 | 32.7 | 129.9 | 2,515 | 15,219.8 | 502.8 | 3,700.2 | 4,051.6 | 18,065.6 | 1,898.4 | 14.2 | 3,124.9 | 6,966.4 | 24,644.2 | 753.1 | 4,577.9 | 51.2 | 105.6 | |
2021.10 | 1,003.8 | 89.4 | 820.4 | 313.8 | 485.7 | 12,457.5 | 20.3 | 5.4 | 921.4 | 39,501.6 | 180.5 | 250.4 | 32.4 | 124.5 | 1,819.1 | 14,109.7 | 451 | 2,912.9 | 3,892.1 | 14,452.1 | 1,733.1 | 10.2 | 2,599.1 | 6,469 | 21,470.9 | 471.3 | 4,147.1 | 47.4 | 79 | |
2021.10 | 973.3 | 118.7 | 687.6 | 212.4 | 486.1 | 11,107.5 | 20.5 | 1 | 1,016.1 | 44,523.2 | 173.1 | 119.6 | 32.5 | 139.4 | 1,481 | 14,896.6 | 381.8 | 2,824 | 3,656.4 | 11,405.6 | 1,711 | 6.8 | 2,670.6 | 5,748.6 | 12,417.1 | 330.1 | 4,159.6 | 42 | 69.5 | |
2021.10 | 954.1 | 95.8 | 793.8 | 292.2 | 442.1 | 11,921.2 | 18.8 | 5.7 | 874.2 | 45,372.9 | 157.2 | 241.4 | 32.3 | 116.3 | 1,713.9 | 11,923.1 | 419 | 2,781.5 | 3,634.3 | 15,074.8 | 1,432.9 | 3.2 | 2,681.8 | 5,438.4 | 18,381.5 | 454.4 | 4,221.1 | 44.1 | 73.2 | |
2021.10 | 926.4 | 110.1 | 815 | 345.4 | 508.4 | 12,025 | 32.2 | 4.8 | 919.4 | 38,345.6 | 167.8 | 341.8 | 32.7 | 128.2 | 1,818.2 | 14,711.4 | 473.8 | 3,217.1 | 3,832.1 | 15,753 | 1,446.4 | 11.5 | 2,720.6 | 6,140.3 | 22,265.4 | 486.8 | 4,789.2 | 49.8 | 91.7 | |
2021.10 | 820 | 74.9 | 683.3 | 643.2 | 653.5 | 6,370 | 34.8 | 0.5 | 760.6 | 22,616.8 | 171.3 | 143.1 | 33.1 | 53.3 | 1,404.5 | 6,567.8 | 387.2 | 1,670.5 | 3,090.8 | 13,394.9 | 1,084.2 | -2.7 | 305.3 | 5,138 | 11,834.1 | 271.2 | 2,631.1 | 36.2 | 71.7 | |
2021.10 | 806.5 | 98 | 828.9 | 135.5 | 367.2 | 9,180 | 38.3 | 2 | 936.4 | 41,924 | 241.8 | 316.4 | 33 | 121.6 | 1,995 | 12,816 | 428.4 | 1,923.9 | 3,788.7 | 13,389.9 | 1,223.5 | -0.1 | 3,969.2 | 3,921.4 | 16,210 | 1,016.8 | 3,880.2 | 47.6 | 88.1 | |
2021.10 | 82.5 | 12 | 0 | 650 | 0 | 1,468.8 | -43 | 0 | 1,077.2 | 112.5 | 0 | 3.9 | 21.4 | 80 | 0 | 346.2 | 0 | 122.8 | 0.1 | 0 | 105.3 | -21 | -1,000 | 125 | 1,022.9 | 172.5 | 20 | -1.5 | -45.9 |