Atari Games On Atari 2600 James Bond
評価指標
Score
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Score |
|---|---|
| dueling-network-architectures-for-deep | 1312.5 |
| prioritized-experience-replay | 3961.0 |
| recurrent-rational-networks | 1122 |
| recurrent-experience-replay-in-distributed | 25354.0 |
| mastering-atari-with-discrete-world-models-1 | 40445 |
| recurrent-rational-networks | 1137 |
| evolving-simple-programs-for-playing-atari | 6130 |
| dueling-network-architectures-for-deep | 835.5 |
| prioritized-experience-replay | 5148.0 |
| the-arcade-learning-environment-an-evaluation | 202.8 |
| generalized-data-distribution-iteration | 594500 |
| self-imitation-learning | 310.8 |
| distributed-prioritized-experience-replay | 21322.5 |
| implicit-quantile-networks-for-distributional | 35108 |
| dna-proximal-policy-optimization-with-a-dual | 14102 |
| agent57-outperforming-the-atari-human | 135784.96 |
| dueling-network-architectures-for-deep | 812.0 |
| impala-scalable-distributed-deep-rl-with | 601.50 |
| soft-actor-critic-for-discrete-action | 68.3 |
| gdi-rethinking-what-makes-reinforcement | 594500 |
| massively-parallel-methods-for-deep | 444.0 |
| online-and-offline-reinforcement-learning-by | 28626.23 |
| fully-parameterized-quantile-function-for | 87291.7 |
| deep-reinforcement-learning-with-double-q | 697.5 |
| deep-reinforcement-learning-with-double-q | 768.5 |
| asynchronous-methods-for-deep-reinforcement | 351.5 |
| train-a-real-world-local-path-planner-in-one | 2237 |
| asynchronous-methods-for-deep-reinforcement | 613.0 |
| learning-values-across-many-orders-of | 507.5 |
| increasing-the-action-gap-new-operators-for | 772.09 |
| human-level-control-through-deep | 576.7 |
| deep-exploration-via-bootstrapped-dqn | 1663.5 |
| mastering-atari-go-chess-and-shogi-by | 41063.25 |
| asynchronous-methods-for-deep-reinforcement | 541.0 |
| policy-optimization-with-penalized-point | 358.54 |
| the-arcade-learning-environment-an-evaluation | 330 |
| distributional-reinforcement-learning-with-1 | 4703 |
| a-distributional-perspective-on-reinforcement | 1909.0 |
| dueling-network-architectures-for-deep | 1358.0 |
| deep-reinforcement-learning-with-double-q | 585.0 |
| モデル 41 | 354.1 |
| deep-reinforcement-learning-with-double-q | 573.0 |
| generalized-data-distribution-iteration | 620780 |
| increasing-the-action-gap-new-operators-for | 848.46 |
| curl-contrastive-unsupervised-representations | - |