{"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/a-study-of-reinforcement-learning-for-neural","title":"A Study of Reinforcement Learning for Neural Machine Translation","arxiv_id":"1808.08866","date":"2018-08-27","proceeding":"EMNLP 2018 10","authors":["Lijun Wu","Fei Tian","Tao Qin","Jian-Huang Lai","Tie-Yan Liu"],"abstract":"Recent studies have shown that reinforcement learning (RL) is an effective\napproach for improving the performance of neural machine translation (NMT)\nsystem. However, due to its instability, successfully RL training is\nchallenging, especially in real-world systems where deep models and large\ndatasets are leveraged. In this paper, taking several large-scale translation\ntasks as testbeds, we conduct a systematic study on how to train better NMT\nmodels using reinforcement learning. We provide a comprehensive comparison of\nseveral important factors (e.g., baseline reward, reward shaping) in RL\ntraining. Furthermore, to fill in the gap that it remains unclear whether RL is\nstill beneficial when monolingual data is used, we propose a new method to\nleverage RL to further boost the performance of NMT systems trained with\nsource/target monolingual data. By integrating all our findings, we obtain\ncompetitive results on WMT14 English- German, WMT17 English-Chinese, and WMT17\nChinese-English translation tasks, especially setting a state-of-the-art\nperformance on WMT17 Chinese-English translation task.","url_abs":"http://arxiv.org/abs/1808.08866v1","url_pdf":"http://arxiv.org/pdf/1808.08866v1.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":"a-study-of-reinforcement-learning-for-neural","repo_url":"https://github.com/apeterswu/RL4NMT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"nmt","task_name":"NMT"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1808.08866","atlas_url":"https://app.syntology.ai/?focus=1808.08866","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1808.08866"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/apeterswu/RL4NMT","reach":null}],"summary":{"ran_honours":1,"ran_draft_wrong":2,"ran_violates":1},"by_repo_kind":{"official":{"samples":4,"ran":4,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":4,"samples":[{"code_sha256_prefix":"0479f465eb03e39f","entry":"compute_bleu","repo":"apeterswu/RL4NMT","repo_kind":"official","path":"tensor2tensor/utils/bleu_hook.py","file_url":"https://github.com/apeterswu/RL4NMT/blob/HEAD/tensor2tensor/utils/bleu_hook.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0479f465eb03e39f"}},{"code_sha256_prefix":"ce9e0ccf3cd0ed9a","entry":"compute_sentence_bleu","repo":"apeterswu/RL4NMT","repo_kind":"official","path":"tensor2tensor/utils/bleu_hook.py","file_url":"https://github.com/apeterswu/RL4NMT/blob/HEAD/tensor2tensor/utils/bleu_hook.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ce9e0ccf3cd0ed9a"}},{"code_sha256_prefix":"3ec104bef4079c4c","entry":"compute_sentence_total_bleu","repo":"apeterswu/RL4NMT","repo_kind":"official","path":"tensor2tensor/utils/bleu_hook.py","file_url":"https://github.com/apeterswu/RL4NMT/blob/HEAD/tensor2tensor/utils/bleu_hook.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3ec104bef4079c4c"}},{"code_sha256_prefix":"c9bd34aafda286a5","entry":"is_class_modality","repo":"apeterswu/RL4NMT","repo_kind":"official","path":"tensor2tensor/utils/t2t_model.py","file_url":"https://github.com/apeterswu/RL4NMT/blob/HEAD/tensor2tensor/utils/t2t_model.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c9bd34aafda286a5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}