{"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/from-language-to-programs-bridging","title":"From Language to Programs: Bridging Reinforcement Learning and Maximum Marginal Likelihood","arxiv_id":"1704.07926","date":"2017-04-25","proceeding":"ACL 2017 7","authors":["Kelvin Guu","Panupong Pasupat","Evan Zheran Liu","Percy Liang"],"abstract":"Our goal is to learn a semantic parser that maps natural language utterances\ninto executable programs when only indirect supervision is available: examples\nare labeled with the correct execution result, but not the program itself.\nConsequently, we must search the space of programs for those that output the\ncorrect result, while not being misled by spurious programs: incorrect programs\nthat coincidentally output the correct result. We connect two common learning\nparadigms, reinforcement learning (RL) and maximum marginal likelihood (MML),\nand then present a new learning algorithm that combines the strengths of both.\nThe new algorithm guards against spurious programs by combining the systematic\nsearch traditionally employed in MML with the randomized exploration of RL, and\nby updating parameters such that probability is spread more evenly across\nconsistent programs. We apply our learning algorithm to a new neural semantic\nparser and show significant gains over existing state-of-the-art results on a\nrecent context-dependent semantic parsing task.","url_abs":"http://arxiv.org/abs/1704.07926v1","url_pdf":"http://arxiv.org/pdf/1704.07926v1.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":"from-language-to-programs-bridging","repo_url":"https://github.com/kelvinguu/lang2program","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"from-language-to-programs-bridging","repo_url":"https://worksheets.codalab.org/worksheets/0x88c914ee1d4b4a4587a07f36f090f3e5","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null},{"paper_slug":"from-language-to-programs-bridging","repo_url":"https://gitlab.com/tau-nlp/vbsix-lang2program","is_official":0,"mentioned_in_paper":0,"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":"semantic-parsing","task_name":"Semantic Parsing"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.07926","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}