{"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/recurrent-dropout-without-memory-loss","title":"Recurrent Dropout without Memory Loss","arxiv_id":"1603.05118","date":"2016-03-16","proceeding":"COLING 2016 12","authors":["Stanislau Semeniuta","Aliaksei Severyn","Erhardt Barth"],"abstract":"This paper presents a novel approach to recurrent neural network (RNN)\nregularization. Differently from the widely adopted dropout method, which is\napplied to \\textit{forward} connections of feed-forward architectures or RNNs,\nwe propose to drop neurons directly in \\textit{recurrent} connections in a way\nthat does not cause loss of long-term memory. Our approach is as easy to\nimplement and apply as the regular feed-forward dropout and we demonstrate its\neffectiveness for Long Short-Term Memory network, the most popular type of RNN\ncells. Our experiments on NLP benchmarks show consistent improvements even when\ncombined with conventional feed-forward dropout.","url_abs":"http://arxiv.org/abs/1603.05118v2","url_pdf":"http://arxiv.org/pdf/1603.05118v2.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":"recurrent-dropout-without-memory-loss","repo_url":"https://github.com/stas-semeniuta/drop-rnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"recurrent-dropout-without-memory-loss","repo_url":"https://github.com/daehwannam/pytorch-rnn-util","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"recurrent-dropout","method_name":"Recurrent Dropout"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1603.05118","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}