Atari Games On Atari 2600 Space Invaders
Metrics
Score
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Score |
|---|---|
| asynchronous-methods-for-deep-reinforcement | 2214.7 |
| evolution-strategies-as-a-scalable | 678.5 |
| prioritized-experience-replay | 2865.8 |
| mastering-atari-with-discrete-world-models-1 | 2474 |
| playing-atari-with-six-neurons | 830 |
| policy-optimization-with-penalized-point | 1216.15 |
| gdi-rethinking-what-makes-reinforcement | 140460 |
| learning-values-across-many-orders-of | 2589.7 |
| generalized-data-distribution-iteration | 140460 |
| fully-parameterized-quantile-function-for | 46498.3 |
| deep-reinforcement-learning-with-double-q | 1692.3 |
| massively-parallel-methods-for-deep | 1183.3 |
| noisy-networks-for-exploration | 5909 |
| asynchronous-methods-for-deep-reinforcement | 15730.5 |
| asynchronous-methods-for-deep-reinforcement | 23846.0 |
| playing-atari-with-deep-reinforcement | 1075 |
| dna-proximal-policy-optimization-with-a-dual | 2731 |
| deep-reinforcement-learning-with-double-q | 8978.0 |
| generalized-data-distribution-iteration | 154380 |
| recurrent-experience-replay-in-distributed | 43223.4 |
| model-free-episodic-control-with-state | 1990 |
| mastering-atari-go-chess-and-shogi-by | 74335.30 |
| Model 23 | 267.9 |
| recurrent-rational-networks | 1395 |
| implicit-quantile-networks-for-distributional | 28888 |
| online-and-offline-reinforcement-learning-by | 3645.63 |
| deep-reinforcement-learning-with-double-q | 1293.8 |
| soft-actor-critic-for-discrete-action | 160.8 |
| the-arcade-learning-environment-an-evaluation | 250.1 |
| impala-scalable-distributed-deep-rl-with | 43595.78 |
| dueling-network-architectures-for-deep | 5993.1 |
| generalized-data-distribution-iteration | 154380 |
| agent57-outperforming-the-atari-human | 48680.86 |
| increasing-the-action-gap-new-operators-for | 3460.79 |
| mean-actor-critic | 1173.1 |
| human-level-control-through-deep | 1976.0 |
| distributed-deep-reinforcement-learning-learn | 650 |
| deep-reinforcement-learning-with-double-q | 2628.7 |
| deep-exploration-via-bootstrapped-dqn | 2893 |
| dueling-network-architectures-for-deep | 15311.5 |
| deep-attention-recurrent-q-network | 650 |
| dueling-network-architectures-for-deep | 2525.5 |
| recurrent-rational-networks | 650 |
| increasing-the-action-gap-new-operators-for | 3277.59 |
| self-imitation-learning | 2951.7 |
| evolving-simple-programs-for-playing-atari | 1001 |
| the-arcade-learning-environment-an-evaluation | 2718 |
| distributional-reinforcement-learning-with-1 | 20972 |
| dueling-network-architectures-for-deep | 6427.3 |
| train-a-real-world-local-path-planner-in-one | 21602 |
| iq-learn-inverse-soft-q-learning-for | - |
| distributed-prioritized-experience-replay | 54681 |
| prioritized-experience-replay | 3912.1 |
| a-distributional-perspective-on-reinforcement | 5747.0 |
| rainbow-combining-improvements-in-deep | 12629.0 |