{"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/hybrid-reward-architecture-for-reinforcement","title":"Hybrid Reward Architecture for Reinforcement Learning","arxiv_id":"1706.04208","date":"2017-06-13","proceeding":"NeurIPS 2017 12","authors":["Harm van Seijen","Mehdi Fatemi","Joshua Romoff","Romain Laroche","Tavian Barnes","Jeffrey Tsang"],"abstract":"One of the main challenges in reinforcement learning (RL) is generalisation.\nIn typical deep RL methods this is achieved by approximating the optimal value\nfunction with a low-dimensional representation using a deep network. While this\napproach works well in many domains, in domains where the optimal value\nfunction cannot easily be reduced to a low-dimensional representation, learning\ncan be very slow and unstable. This paper contributes towards tackling such\nchallenging domains, by proposing a new method, called Hybrid Reward\nArchitecture (HRA). HRA takes as input a decomposed reward function and learns\na separate value function for each component reward function. Because each\ncomponent typically only depends on a subset of all features, the corresponding\nvalue function can be approximated more easily by a low-dimensional\nrepresentation, enabling more effective learning. We demonstrate HRA on a\ntoy-problem and the Atari game Ms. Pac-Man, where HRA achieves above-human\nperformance.","url_abs":"http://arxiv.org/abs/1706.04208v2","url_pdf":"http://arxiv.org/pdf/1706.04208v2.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":"hybrid-reward-architecture-for-reinforcement","repo_url":"https://github.com/KhenNguyn/DoAn3-MachineLearning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.04208","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}