{"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/learning-goal-embeddings-via-self-play-for","title":"Learning Goal Embeddings via Self-Play for Hierarchical Reinforcement Learning","arxiv_id":"1811.09083","date":"2018-11-22","proceeding":null,"authors":["Sainbayar Sukhbaatar","Emily Denton","Arthur Szlam","Rob Fergus"],"abstract":"In hierarchical reinforcement learning a major challenge is determining\nappropriate low-level policies. We propose an unsupervised learning scheme,\nbased on asymmetric self-play from Sukhbaatar et al. (2018), that automatically\nlearns a good representation of sub-goals in the environment and a low-level\npolicy that can execute them. A high-level policy can then direct the lower one\nby generating a sequence of continuous sub-goal vectors. We evaluate our model\nusing Mazebase and Mujoco environments, including the challenging AntGather\ntask. Visualizations of the sub-goal embeddings reveal a logical decomposition\nof tasks within the environment. Quantitatively, our approach obtains\ncompelling performance gains over non-hierarchical approaches.","url_abs":"http://arxiv.org/abs/1811.09083v1","url_pdf":"http://arxiv.org/pdf/1811.09083v1.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":"learning-goal-embeddings-via-self-play-for","repo_url":"https://github.com/ACampero/hsp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"learning-goal-embeddings-via-self-play-for","repo_url":"https://github.com/tesatory/hsp","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"hierarchical-reinforcement-learning","task_name":"Hierarchical Reinforcement Learning"},{"task_slug":"mujoco","task_name":"MuJoCo"},{"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":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1811.09083","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}