Atari Games On Atari 2600 Chopper Command
评估指标
Score
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | Score |
|---|---|
| self-imitation-learning | 6710 |
| increasing-the-action-gap-new-operators-for | 5431.36 |
| gdi-rethinking-what-makes-reinforcement | 999999 |
| evolving-simple-programs-for-playing-atari | 3580 |
| prioritized-experience-replay | 4635.0 |
| noisy-networks-for-exploration | 11477 |
| recurrent-experience-replay-in-distributed | 986652.0 |
| policy-optimization-with-penalized-point | 6308.33 |
| distributed-prioritized-experience-replay | 721851 |
| deep-reinforcement-learning-with-double-q | 6126.0 |
| deep-reinforcement-learning-with-double-q | 8058.0 |
| deep-reinforcement-learning-with-double-q | 5017.0 |
| mastering-atari-go-chess-and-shogi-by | 991039.70 |
| implicit-quantile-networks-for-distributional | 16836 |
| massively-parallel-methods-for-deep | 3191.8 |
| the-reactor-a-fast-and-sample-efficient-actor | 107779.0 |
| dna-proximal-policy-optimization-with-a-dual | 31181 |
| fully-parameterized-quantile-function-for | 876460.0 |
| asynchronous-methods-for-deep-reinforcement | 10150.0 |
| a-distributional-perspective-on-reinforcement | 15600.0 |
| increasing-the-action-gap-new-operators-for | 5734.93 |
| dueling-network-architectures-for-deep | 11215.0 |
| mastering-atari-with-discrete-world-models-1 | 2861 |
| agent57-outperforming-the-atari-human | 999900 |
| online-and-offline-reinforcement-learning-by | 5989.55 |
| prioritized-experience-replay | 8600.0 |
| learning-values-across-many-orders-of | 775.0 |
| asynchronous-methods-for-deep-reinforcement | 4669.0 |
| distributional-reinforcement-learning-with-1 | 14667 |
| generalized-data-distribution-iteration | 999999 |
| asynchronous-methods-for-deep-reinforcement | 7021.0 |
| generalized-data-distribution-iteration | 999999 |
| deep-reinforcement-learning-with-double-q | 3495.0 |
| impala-scalable-distributed-deep-rl-with | 28255.00 |
| human-level-control-through-deep | 6687.0 |
| train-a-real-world-local-path-planner-in-one | 15071 |
| deep-exploration-via-bootstrapped-dqn | 4100 |
| dueling-network-architectures-for-deep | 5809.0 |
| the-arcade-learning-environment-an-evaluation | 34018.8 |
| curl-contrastive-unsupervised-representations | 1198 |
| evolution-strategies-as-a-scalable | 3710.0 |
| the-arcade-learning-environment-an-evaluation | 1581.5 |
| dueling-network-architectures-for-deep | 13185.0 |
| 模型 44 | 16.9 |
| dueling-network-architectures-for-deep | 3784.0 |