{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/state-of-the-art-control-of-atari-games-using","title":"State of the Art Control of Atari Games Using Shallow Reinforcement Learning","arxiv_id":"1512.01563","date":"2015-12-04","proceeding":null,"authors":["Yitao Liang","Marlos C. Machado","Erik Talvitie","Michael Bowling"],"abstract":"The recently introduced Deep Q-Networks (DQN) algorithm has gained attention\nas one of the first successful combinations of deep neural networks and\nreinforcement learning. Its promise was demonstrated in the Arcade Learning\nEnvironment (ALE), a challenging framework composed of dozens of Atari 2600\ngames used to evaluate general competency in AI. It achieved dramatically\nbetter results than earlier approaches, showing that its ability to learn good\nrepresentations is quite robust and general. This paper attempts to understand\nthe principles that underlie DQN's impressive performance and to better\ncontextualize its success. We systematically evaluate the importance of key\nrepresentational biases encoded by DQN's network by proposing simple linear\nrepresentations that make use of these concepts. Incorporating these\ncharacteristics, we obtain a computationally practical feature set that\nachieves competitive performance to DQN in the ALE. Besides offering insight\ninto the strengths and weaknesses of DQN, we provide a generic representation\nfor the ALE, significantly reducing the burden of learning a representation for\neach game. Moreover, we also provide a simple, reproducible benchmark for the\nsake of comparison to future work in the ALE.","url_abs":"http://arxiv.org/abs/1512.01563v2","url_pdf":"http://arxiv.org/pdf/1512.01563v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"state-of-the-art-control-of-atari-games-using","repo_url":"https://github.com/mcmachado/b-pro","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"atari-games","task_name":"Atari Games"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dqn","method_name":"DQN"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"q-learning","method_name":"Q-Learning"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1512.01563","atlas_url":"https://app.syntology.ai/?focus=1512.01563","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}