{"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/constrained-exploration-and-recovery-from","title":"Constrained Exploration and Recovery from Experience Shaping","arxiv_id":"1809.08925","date":"2018-09-21","proceeding":null,"authors":["Tu-Hoa Pham","Giovanni De Magistris","Don Joven Agravante","Subhajit Chaudhury","Asim Munawar","Ryuki Tachibana"],"abstract":"We consider the problem of reinforcement learning under safety requirements,\nin which an agent is trained to complete a given task, typically formalized as\nthe maximization of a reward signal over time, while concurrently avoiding\nundesirable actions or states, associated to lower rewards, or penalties. The\nconstruction and balancing of different reward components can be difficult in\nthe presence of multiple objectives, yet is crucial for producing a satisfying\npolicy. For example, in reaching a target while avoiding obstacles, low\ncollision penalties can lead to reckless movements while high penalties can\ndiscourage exploration. To circumvent this limitation, we examine the effect of\npast actions in terms of safety to estimate which are acceptable or should be\navoided in the future. We then actively reshape the action space of the agent\nduring reinforcement learning, so that reward-driven exploration is constrained\nwithin safety limits. We propose an algorithm enabling the learning of such\nsafety constraints in parallel with reinforcement learning and demonstrate its\neffectiveness in terms of both task completion and training time.","url_abs":"http://arxiv.org/abs/1809.08925v1","url_pdf":"http://arxiv.org/pdf/1809.08925v1.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":"constrained-exploration-and-recovery-from","repo_url":"https://github.com/IBM/constrained-rl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"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":{"syntology_url":"https://syntology.ai/paper/1809.08925","atlas_url":"https://app.syntology.ai/?focus=1809.08925","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}