Atari Games On Atari 2600 Asterix
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 |
|---|---|
| distributed-prioritized-experience-replay | 313305 |
| モデル 2 | 1332 |
| evolution-strategies-as-a-scalable | 1440 |
| prioritized-experience-replay | 22484.5 |
| dueling-network-architectures-for-deep | 364200.0 |
| evolving-simple-programs-for-playing-atari | 1880 |
| deep-reinforcement-learning-with-double-q | 16837.0 |
| noisy-networks-for-exploration | 28350 |
| soft-actor-critic-for-discrete-action | 272 |
| dna-proximal-policy-optimization-with-a-dual | 323965 |
| recurrent-experience-replay-in-distributed | 999153.3 |
| train-a-real-world-local-path-planner-in-one | 567640 |
| the-arcade-learning-environment-an-evaluation | 987.3 |
| learning-values-across-many-orders-of | 18919.5 |
| impala-scalable-distributed-deep-rl-with | 300732.00 |
| human-level-control-through-deep | 6012 |
| asynchronous-methods-for-deep-reinforcement | 22140.5 |
| generalized-data-distribution-iteration | 999999 |
| dueling-network-architectures-for-deep | 17356.5 |
| dueling-network-architectures-for-deep | 375080.0 |
| online-and-offline-reinforcement-learning-by | 862406.65 |
| curl-contrastive-unsupervised-representations | 524.3 |
| distributional-reinforcement-learning-with-1 | 261025 |
| prioritized-experience-replay | 31527 |
| recurrent-rational-networks | 12621 |
| fully-parameterized-quantile-function-for | 578388.5 |
| massively-parallel-methods-for-deep | 3324.7 |
| generalized-data-distribution-iteration | 759910 |
| dueling-network-architectures-for-deep | 15840.0 |
| deep-reinforcement-learning-with-double-q | 3170.5 |
| モデル 31 | 21040 |
| increasing-the-action-gap-new-operators-for | 19564.9 |
| implicit-quantile-networks-for-distributional | 342016 |
| recurrent-rational-networks | 18109 |
| self-imitation-learning | 17984.2 |
| a-distributional-perspective-on-reinforcement | 406211 |
| deep-exploration-via-bootstrapped-dqn | 19713.2 |
| mastering-atari-with-discrete-world-models-1 | 72311 |
| asynchronous-methods-for-deep-reinforcement | 17244.5 |
| the-reactor-a-fast-and-sample-efficient-actor | 205914.0 |
| deep-reinforcement-learning-with-double-q | 364200.0 |
| deep-reinforcement-learning-with-double-q | 4359.0 |
| the-arcade-learning-environment-an-evaluation | 290700 |
| mastering-atari-go-chess-and-shogi-by | 998425.00 |
| asynchronous-methods-for-deep-reinforcement | 6723 |
| dueling-network-architectures-for-deep | 28188.0 |
| agent57-outperforming-the-atari-human | 991384.42 |
| policy-optimization-with-penalized-point | 4310.67 |
| increasing-the-action-gap-new-operators-for | 12852.08 |