{"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/training-language-models-using-target","title":"Training Language Models Using Target-Propagation","arxiv_id":"1702.04770","date":"2017-02-15","proceeding":null,"authors":["Sam Wiseman","Sumit Chopra","Marc'Aurelio Ranzato","Arthur Szlam","Ruoyu Sun","Soumith Chintala","Nicolas Vasilache"],"abstract":"While Truncated Back-Propagation through Time (BPTT) is the most popular\napproach to training Recurrent Neural Networks (RNNs), it suffers from being\ninherently sequential (making parallelization difficult) and from truncating\ngradient flow between distant time-steps. We investigate whether Target\nPropagation (TPROP) style approaches can address these shortcomings.\nUnfortunately, extensive experiments suggest that TPROP generally underperforms\nBPTT, and we end with an analysis of this phenomenon, and suggestions for\nfuture work.","url_abs":"http://arxiv.org/abs/1702.04770v1","url_pdf":"http://arxiv.org/pdf/1702.04770v1.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":"training-language-models-using-target","repo_url":"https://github.com/facebookresearch/TPRNN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1702.04770","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}