{"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/reinforcement-learning-with-unsupervised","title":"Reinforcement Learning with Unsupervised Auxiliary Tasks","arxiv_id":"1611.05397","date":"2016-11-16","proceeding":null,"authors":["Max Jaderberg","Volodymyr Mnih","Wojciech Marian Czarnecki","Tom Schaul","Joel Z. Leibo","David Silver","Koray Kavukcuoglu"],"abstract":"Deep reinforcement learning agents have achieved state-of-the-art results by\ndirectly maximising cumulative reward. However, environments contain a much\nwider variety of possible training signals. In this paper, we introduce an\nagent that also maximises many other pseudo-reward functions simultaneously by\nreinforcement learning. All of these tasks share a common representation that,\nlike unsupervised learning, continues to develop in the absence of extrinsic\nrewards. We also introduce a novel mechanism for focusing this representation\nupon extrinsic rewards, so that learning can rapidly adapt to the most relevant\naspects of the actual task. Our agent significantly outperforms the previous\nstate-of-the-art on Atari, averaging 880\\% expert human performance, and a\nchallenging suite of first-person, three-dimensional \\emph{Labyrinth} tasks\nleading to a mean speedup in learning of 10$\\times$ and averaging 87\\% expert\nhuman performance on Labyrinth.","url_abs":"http://arxiv.org/abs/1611.05397v1","url_pdf":"http://arxiv.org/pdf/1611.05397v1.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":"reinforcement-learning-with-unsupervised","repo_url":"https://github.com/OliverRichter/map-reader","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"reinforcement-learning-with-unsupervised","repo_url":"https://github.com/miyosuda/unreal","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"reinforcement-learning-with-unsupervised","repo_url":"https://github.com/takuseno/unreal","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"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":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1611.05397","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}