{"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/grounding-language-for-transfer-in-deep","title":"Grounding Language for Transfer in Deep Reinforcement Learning","arxiv_id":"1708.00133","date":"2017-08-01","proceeding":null,"authors":["Karthik Narasimhan","Regina Barzilay","Tommi Jaakkola"],"abstract":"In this paper, we explore the utilization of natural language to drive\ntransfer for reinforcement learning (RL). Despite the wide-spread application\nof deep RL techniques, learning generalized policy representations that work\nacross domains remains a challenging problem. We demonstrate that textual\ndescriptions of environments provide a compact intermediate channel to\nfacilitate effective policy transfer. Specifically, by learning to ground the\nmeaning of text to the dynamics of the environment such as transitions and\nrewards, an autonomous agent can effectively bootstrap policy learning on a new\ndomain given its description. We employ a model-based RL approach consisting of\na differentiable planning module, a model-free component and a factorized state\nrepresentation to effectively use entity descriptions. Our model outperforms\nprior work on both transfer and multi-task scenarios in a variety of different\nenvironments. For instance, we achieve up to 14% and 11.5% absolute improvement\nover previously existing models in terms of average and initial rewards,\nrespectively.","url_abs":"http://arxiv.org/abs/1708.00133v2","url_pdf":"http://arxiv.org/pdf/1708.00133v2.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":"grounding-language-for-transfer-in-deep","repo_url":"https://github.com/karthikncode/Grounded-RL-Transfer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":null}],"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":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}