{"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/unsupervised-visuomotor-control-through","title":"Unsupervised Visuomotor Control through Distributional Planning Networks","arxiv_id":"1902.05542","date":"2019-02-14","proceeding":null,"authors":["Tianhe Yu","Gleb Shevchuk","Dorsa Sadigh","Chelsea Finn"],"abstract":"While reinforcement learning (RL) has the potential to enable robots to\nautonomously acquire a wide range of skills, in practice, RL usually requires\nmanual, per-task engineering of reward functions, especially in real world\nsettings where aspects of the environment needed to compute progress are not\ndirectly accessible. To enable robots to autonomously learn skills, we instead\nconsider the problem of reinforcement learning without access to rewards. We\naim to learn an unsupervised embedding space under which the robot can measure\nprogress towards a goal for itself. Our approach explicitly optimizes for a\nmetric space under which action sequences that reach a particular state are\noptimal when the goal is the final state reached. This enables learning\neffective and control-centric representations that lead to more autonomous\nreinforcement learning algorithms. Our experiments on three simulated\nenvironments and two real-world manipulation problems show that our method can\nlearn effective goal metrics from unlabeled interaction, and use the learned\ngoal metrics for autonomous reinforcement learning.","url_abs":"http://arxiv.org/abs/1902.05542v1","url_pdf":"http://arxiv.org/pdf/1902.05542v1.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":"unsupervised-visuomotor-control-through","repo_url":"https://github.com/tianheyu927/dpn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"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=1902.05542","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}