Atari Games On Atari 2600 Bowling
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 |
|---|---|
| increasing-the-action-gap-new-operators-for | 57.41 |
| gdi-rethinking-what-makes-reinforcement | 201.9 |
| deep-reinforcement-learning-with-double-q | 69.6 |
| dna-proximal-policy-optimization-with-a-dual | 181 |
| asynchronous-methods-for-deep-reinforcement | 41.8 |
| evolving-simple-programs-for-playing-atari | 85.8 |
| generalized-data-distribution-iteration | 205.2 |
| impala-scalable-distributed-deep-rl-with | 59.92 |
| dueling-network-architectures-for-deep | 65.5 |
| train-a-real-world-local-path-planner-in-one | 62.4 |
| deep-reinforcement-learning-with-double-q | 50.4 |
| distributed-prioritized-experience-replay | 17.6 |
| implicit-quantile-networks-for-distributional | 86.5 |
| dueling-network-architectures-for-deep | 65.7 |
| distributional-reinforcement-learning-with-1 | 77.2 |
| asynchronous-methods-for-deep-reinforcement | 35.1 |
| dueling-network-architectures-for-deep | 68.1 |
| rudder-return-decomposition-for-delayed | 179 |
| increasing-the-action-gap-new-operators-for | 71.59 |
| massively-parallel-methods-for-deep | 54 |
| self-imitation-learning | 31.1 |
| prioritized-experience-replay | 52 |
| policy-optimization-with-penalized-point | 38.99 |
| mastering-atari-with-discrete-world-models-1 | 49 |
| generalized-data-distribution-iteration | 201.9 |
| first-return-then-explore | 260 |
| dueling-network-architectures-for-deep | 46.7 |
| fully-parameterized-quantile-function-for | 102.3 |
| deep-reinforcement-learning-with-double-q | 50.4 |
| deep-reinforcement-learning-with-double-q | 56.5 |
| a-distributional-perspective-on-reinforcement | 81.8 |
| online-and-offline-reinforcement-learning-by | 131.65 |
| learning-values-across-many-orders-of | 102.1 |
| the-reactor-a-fast-and-sample-efficient-actor | 81.0 |
| evolution-strategies-as-a-scalable | 30 |
| asynchronous-methods-for-deep-reinforcement | 36.2 |
| agent57-outperforming-the-atari-human | 251.18 |
| prioritized-experience-replay | 47.9 |
| deep-exploration-via-bootstrapped-dqn | 60.2 |
| human-level-control-through-deep | 42.4 |
| mastering-atari-go-chess-and-shogi-by | 260.13 |
| モデル 42 | 36.4 |
| recurrent-experience-replay-in-distributed | 219.5 |
| the-arcade-learning-environment-an-evaluation | 43.9 |