{"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/gradual-learning-of-recurrent-neural-networks","title":"Gradual Learning of Recurrent Neural Networks","arxiv_id":"1708.08863","date":"2017-08-29","proceeding":null,"authors":["Ziv Aharoni","Gal Rattner","Haim Permuter"],"abstract":"Recurrent Neural Networks (RNNs) achieve state-of-the-art results in many\nsequence-to-sequence modeling tasks. However, RNNs are difficult to train and\ntend to suffer from overfitting. Motivated by the Data Processing Inequality\n(DPI), we formulate the multi-layered network as a Markov chain, introducing a\ntraining method that comprises training the network gradually and using\nlayer-wise gradient clipping. We found that applying our methods, combined with\npreviously introduced regularization and optimization methods, resulted in\nimprovements in state-of-the-art architectures operating in language modeling\ntasks.","url_abs":"http://arxiv.org/abs/1708.08863v2","url_pdf":"http://arxiv.org/pdf/1708.08863v2.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":"gradual-learning-of-recurrent-neural-networks","repo_url":"https://github.com/zivaharoni/gradual-learning-rnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/language-modelling-on-penn-treebank-word","task":"Language Modelling","dataset":"Penn Treebank (Word Level)","model":"GL-LWGC + AWD-MoS-LSTM + dynamic eval","rank_in_archive_order":6,"of":43,"metrics":{"Params":"26M","Test perplexity":"46.34","Validation perplexity":"46.64"},"uses_additional_data":false},{"leaderboard":"/sota/language-modelling-on-wikitext-2","task":"Language Modelling","dataset":"WikiText-2","model":"GL-LWGC + AWD-MoS-LSTM + dynamic eval","rank_in_archive_order":15,"of":38,"metrics":{"Number of params":"38M","Test perplexity":"40.46","Validation perplexity":"42.19"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}