mlsanityData to 7 Oct 2026

Agent57: Outperforming the Atari Human Benchmark

arXiv 2003.13350 ↗First posted 30 Mar 2020cs.LG stat.MLcited by 300

Not on the map. Why?

Abstract (arXiv)

Atari games have been a long-standing benchmark in the reinforcement learning (RL) community for the past decade. This benchmark was proposed to test general competency of RL algorithms. Previous work has achieved good average performance by doing outstandingly well on many games of the set, but very poorly in several of the most challenging games. We propose Agent57, the first deep RL agent that outperforms the standard human benchmark on all 57 Atari games. To achieve this result, we train a neural network which parameterizes a family of policies ranging from very exploratory to purely exploitative. We propose an adaptive mechanism to choose which policy to prioritize throughout the training process. Additionally, we utilize a novel parameterization of the architecture that allows for more consistent and stable learning.

Key facts. Cited by 300 papers in the map (data through 7 Oct 2026). First posted on arXiv 30 Mar 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 88; 2022: 70; 2023: 52; 2024: 42; 2025: 30; 2026: 18.

300 papers.

Sort:

By area:

  1. 2021Deep Reinforcement Learning at the Edge of the Statistical Precipicecited by 781
  2. 2022Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videoscited by 338
  3. 2021Benchmarking the Spectrum of Agent Capabilitiescited by 193
  4. 2021Behavior From the Void: Unsupervised Active Pre-Trainingcited by 173
  5. 2022Hybrid RL: Using Both Offline and Online Data Can Make RL Efficientcited by 158
  6. 2021APS: Active Pretraining with Successor Featurescited by 109
  7. 2021Celebrating Diversity in Shared Multi-Agent Reinforcement Learningcited by 96
  8. 2021MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Researchcited by 96
  9. 2022Deep Hierarchical Planning from Pixelscited by 96
  10. 2021Going Beyond Linear Transformers with Recurrent Fast Weight Programmerscited by 84
  11. 2021Neural Production Systems: Learning Rule-Governed Visual Dynamicscited by 75
  12. 2022BYOL-Explore: Exploration by Bootstrapped Predictioncited by 71
38 more
  1. 2024Simplifying Deep Temporal Difference Learningcited by 71
  2. 2022Automated Reinforcement Learning (AutoRL): A Survey and Open Problemscited by 69
  3. 2021Learning to Fly -- a Gym Environment with PyBullet Physics for Reinforcement Learning of Multi-agent Quadcopter Controlcited by 68
  4. 2023Elastic Decision Transformercited by 66
  5. 2022A Unified Approach to Reinforcement Learning, Quantal Response Equilibria, and Two-Player Zero-Sum Gamescited by 64
  6. 2022Lipschitz-constrained Unsupervised Skill Discoverycited by 60
  7. 2021Quantum Architecture Search via Deep Reinforcement Learningcited by 58
  8. 2023Hyperparameters in Reinforcement Learning and How To Tune Themcited by 58
  9. 2021Exploration in Deep Reinforcement Learning: From Single-Agent to Multiagent Domaincited by 57
  10. 2021Tactical Optimism and Pessimism for Deep Reinforcement Learningcited by 55
  11. 2021Learning to run a Power Network Challenge: a Retrospective Analysiscited by 52
  12. 2022Semantic Exploration from Language Abstractions and Pretrained Representationscited by 51
  13. 2021Variational Quantum Reinforcement Learning via Evolutionary Optimizationcited by 50
  14. 2021Episodic Multi-agent Reinforcement Learning with Curiosity-Driven Explorationcited by 49
  15. 2023Anti-Exploration by Random Network Distillationcited by 49
  16. 2021Landmark-Guided Subgoal Generation in Hierarchical Reinforcement Learningcited by 46
  17. 2023Read and Reap the Rewards: Learning to Play Atari with the Help of Instruction Manualscited by 45
  18. 2022Atari-5: Distilling the Arcade Learning Environment down to Five Gamescited by 44
  19. 2021Human-Level Reinforcement Learning through Theory-Based Modeling, Exploration, and Planningcited by 43
  20. 2022Guaranteed Discovery of Control-Endogenous Latent States with Multi-Step Inverse Modelscited by 42
  21. 2022Contextualize Me -- The Case for Context in Reinforcement Learningcited by 41
  22. 2022Evaluating Long-Term Memory in 3D Mazescited by 41
  23. 2025Infrastructure for AI Agentscited by 40
  24. 2025Towards General-Purpose Model-Free Reinforcement Learningcited by 40
  25. 2021Learning more skills through optimistic explorationcited by 37
  26. 2022A Review of Uncertainty for Deep Reinforcement Learningcited by 36
  27. 2023AMAGO: Scalable In-Context Reinforcement Learning for Adaptive Agentscited by 36
  28. 2021Geometric Entropic Explorationcited by 32
  29. 2022Why Should I Trust You, Bellman? The Bellman Error is a Poor Replacement for Value Errorcited by 31
  30. 2021Ensemble Bootstrapping for Q-Learningcited by 30
  31. 2023On the Importance of Exploration for Generalization in Reinforcement Learningcited by 30
  32. 2023OCAtari: Object-Centric Atari 2600 Reinforcement Learning Environmentscited by 30
  33. 2023Multi-agent Reinforcement Learning: A Comprehensive Surveycited by 30
  34. 2024Weight Clipping for Deep Continual and Reinforcement Learningcited by 29
  35. 2022Human-level Atari 200x fastercited by 28
  36. 2023Flipping Coins to Estimate Pseudocounts for Exploration in Reinforcement Learningcited by 28
  37. 2021Synthetic Returns for Long-Term Credit Assignmentcited by 27
  38. 2022The Free Energy Principle for Perception and Action: A Deep Learning Perspectivecited by 24

50 most cited of 300 All 300 (JSON)