Atari Games On Atari 2600 Private Eye
평가 지표
Score
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | Score |
|---|---|
| gdi-rethinking-what-makes-reinforcement | 15100 |
| count-based-exploration-with-the-successor | 99.1 |
| deep-reinforcement-learning-with-double-q | -575.5 |
| first-return-then-explore | 95756 |
| self-imitation-learning | 661.2 |
| deep-reinforcement-learning-with-double-q | 146.7 |
| mastering-atari-with-discrete-world-models-1 | 2198 |
| asynchronous-methods-for-deep-reinforcement | 206.9 |
| a-distributional-perspective-on-reinforcement | 15095.0 |
| recurrent-experience-replay-in-distributed | 5322.7 |
| prioritized-experience-replay | 670.7 |
| generalized-data-distribution-iteration | 15100 |
| generalized-data-distribution-iteration | 15100 |
| distributional-reinforcement-learning-with-1 | 350 |
| 모델 15 | 86.0 |
| count-based-exploration-with-neural-density | 8358.7 |
| dueling-network-architectures-for-deep | 292.6 |
| dueling-network-architectures-for-deep | 206.0 |
| dueling-network-architectures-for-deep | 103.0 |
| prioritized-experience-replay | 200.0 |
| learning-values-across-many-orders-of | 286.7 |
| the-arcade-learning-environment-an-evaluation | 684.3 |
| deep-reinforcement-learning-with-double-q | 1277.6 |
| unifying-count-based-exploration-and | 99.32 |
| exploration-by-self-supervised-exploitation | 15089 |
| impala-scalable-distributed-deep-rl-with | 98.50 |
| evolution-strategies-as-a-scalable | 100.0 |
| asynchronous-methods-for-deep-reinforcement | 421.1 |
| distributed-prioritized-experience-replay | 49.8 |
| policy-optimization-with-penalized-point | 79.67 |
| increasing-the-action-gap-new-operators-for | 5276.16 |
| implicit-quantile-networks-for-distributional | 200 |
| evolving-simple-programs-for-playing-atari | 12702.2 |
| human-level-control-through-deep | 1788.0 |
| dna-proximal-policy-optimization-with-a-dual | 100 |
| deep-reinforcement-learning-with-double-q | 207.9 |
| count-based-exploration-with-neural-density | 206.0 |
| the-arcade-learning-environment-an-evaluation | 1947.3 |
| dueling-network-architectures-for-deep | 129.7 |
| exploration-by-self-supervised-exploitation | 17313 |
| curl-contrastive-unsupervised-representations | 105.2 |
| mastering-atari-go-chess-and-shogi-by | 15299.98 |
| exploration-by-self-supervised-exploitation | 4213 |
| online-and-offline-reinforcement-learning-by | 100 |
| large-scale-study-of-curiosity-driven | 3036.5 |
| massively-parallel-methods-for-deep | 2598.6 |
| train-a-real-world-local-path-planner-in-one | 349.7 |
| agent57-outperforming-the-atari-human | 79716.46 |
| noisy-networks-for-exploration | 279 |
| exploration-by-random-network-distillation | 8666 |
| asynchronous-methods-for-deep-reinforcement | 194.4 |
| deep-exploration-via-bootstrapped-dqn | 1812.5 |