{"id":17267118,"url":"https://github.com/brendanator/atari-rl","last_synced_at":"2025-08-21T01:32:22.916Z","repository":{"id":45297229,"uuid":"78427553","full_name":"brendanator/atari-rl","owner":"brendanator","description":"Atari - Deep Reinforcement Learning algorithms in TensorFlow","archived":false,"fork":false,"pushed_at":"2024-03-27T10:44:40.000Z","size":146,"stargazers_count":135,"open_issues_count":5,"forks_count":34,"subscribers_count":8,"default_branch":"main","last_synced_at":"2025-04-08T01:42:25.075Z","etag":null,"topics":["atari","deep-reinforcement-learning","tensorflow"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/brendanator.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"MIT-LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2017-01-09T12:39:49.000Z","updated_at":"2025-03-21T16:06:15.000Z","dependencies_parsed_at":"2023-01-21T15:02:05.185Z","dependency_job_id":null,"html_url":"https://github.com/brendanator/atari-rl","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/brendanator/atari-rl","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/brendanator%2Fatari-rl","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/brendanator%2Fatari-rl/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/brendanator%2Fatari-rl/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/brendanator%2Fatari-rl/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/brendanator","download_url":"https://codeload.github.com/brendanator/atari-rl/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/brendanator%2Fatari-rl/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":271415042,"owners_count":24755628,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-08-20T02:00:09.606Z","response_time":69,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["atari","deep-reinforcement-learning","tensorflow"],"created_at":"2024-10-15T08:08:53.896Z","updated_at":"2025-08-21T01:32:22.597Z","avatar_url":"https://github.com/brendanator.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Atari - Deep Reinforcement Learning algorithms in TensorFlow\n\n[![Build Status](https://travis-ci.org/brendanator/atari-rl.svg?branch=master)](https://travis-ci.org/brendanator/atari-rl)\n\nLearning to play Atari in TensorFlow using Deep Reinforcement Learning\n\n## Setup\n```\ngit clone https://github.com/brendanator/atari-rl\ngit submodule update --init\nconda create --name atari-rl python=3.5\nsource activate atari-rl\nconda install -y -c https://conda.binstar.org/menpo opencv3\nconda install -y h5py numpy\npip install tensorflow\npip install 'gym[atari]'\n```\nPython 2.7 is also supported\n\n## Usage\n\n- Show all options - `python main.py --help`\n- Play a specific [Atari game](https://github.com/mgbellemare/Arcade-Learning-Environment/blob/master/src/games/Roms.cpp#L17) - `python main.py --game Breakout`\n\n## Papers Implemented\n\n- :white_check_mark: [Human Level Control through Deep Reinforcement Learning](http://www.nature.com/nature/journal/v518/n7540/pdf/nature14236.pdf)\n    - `python main.py`\n- :white_check_mark: [Deep Reinforcement Learning with Double Q-learning](https://arxiv.org/pdf/1509.06461.pdf)\n    - `python main.py --double_q`\n- :white_check_mark: [Dueling Network Architectures for Deep Reinforcement Learning](https://arxiv.org/pdf/1511.06581.pdf)\n    - `python main.py --dueling`\n- :white_check_mark: [Learning to Play in a Day: Faster Deep Reinforcement Learning by Optimality Tightening](https://arxiv.org/pdf/1611.01606.pdf)\n    - `python main.py --optimality_tightening`\n- :white_check_mark: [Prioritized Experience Replay](https://arxiv.org/pdf/1511.05952.pdf)\n    - `python main.py --replay_prioritized`\n    - Only proportional prioritized replay is implemented\n- :white_check_mark: [Unifying Count-Based Exploration and Intrinsic Motivation](https://arxiv.org/pdf/1606.01868.pdf)\n    - `python main.py --exploration_bonus`\n- :white_check_mark: [Deep Exploration via Bootstrapped DQN](https://arxiv.org/pdf/1602.04621.pdf)\n    - `python main.py --bootstrapped`\n- :white_check_mark: [Increasing the Action Gap: New Operators for Reinforcement Learning](https://arxiv.org/pdf/1512.04860.pdf)\n    - `python main.py --persistent_advantage_learning`\n- :white_check_mark: [Learning values across many orders of magnitudes](https://arxiv.org/pdf/1602.07714.pdf)\n    - `python main.py --reward_scaling`\n- :white_check_mark: [Asynchronous Methods for Deep Reinforcement Learning](https://arxiv.org/pdf/1602.01783.pdf)\n    - `python main.py --async one_step`\n    - `python main.py --async n_step`\n    - `python main.py --async n_step --sarsa`\n    - `python main.py --async a3c`\n- :x: [Deep Recurrent Q-Learning for Partially Observable MDPs](https://arxiv.org/pdf/1507.06527.pdf)\n- :x: [Safe and efficient Off-Policy Reinforcement Learning](https://arxiv.org/pdf/1606.02647.pdf)\n- :x: [Continuous Deep Q-Learning with Model-based Acceleration](https://arxiv.org/pdf/1603.00748.pdf)\n\n## Acknowledgements\n\n- https://github.com/mgbellemare/SkipCTS - Used in implementation of [Unifying Count-Based Exploration and Intrinsic Motivation](https://arxiv.org/pdf/1606.01868.pdf)\n- https://github.com/Kaixhin/Atari\n- https://github.com/carpedm20/deep-rl-tensorflow\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbrendanator%2Fatari-rl","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbrendanator%2Fatari-rl","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbrendanator%2Fatari-rl/lists"}