{"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/gating-revisited-deep-multi-layer-rnns-that-1","title":"Gating Revisited: Deep Multi-layer RNNs That Can Be Trained","arxiv_id":"1911.11033","date":"2019-11-25","proceeding":null,"authors":["Mehmet Ozgur Turkoglu","Stefano D'Aronco","Jan Dirk Wegner","Konrad Schindler"],"abstract":"We propose a new STAckable Recurrent cell (STAR) for recurrent neural networks (RNNs), which has fewer parameters than widely used LSTM and GRU while being more robust against vanishing or exploding gradients. Stacking recurrent units into deep architectures suffers from two major limitations: (i) many recurrent cells (e.g., LSTMs) are costly in terms of parameters and computation resources; and (ii) deep RNNs are prone to vanishing or exploding gradients during training. We investigate the training of multi-layer RNNs and examine the magnitude of the gradients as they propagate through the network in the \"vertical\" direction. We show that, depending on the structure of the basic recurrent unit, the gradients are systematically attenuated or amplified. Based on our analysis we design a new type of gated cell that better preserves gradient magnitude. We validate our design on a large number of sequence modelling tasks and demonstrate that the proposed STAR cell allows to build and train deeper recurrent architectures, ultimately leading to improved performance while being computationally more efficient.","url_abs":"https://arxiv.org/abs/1911.11033v4","url_pdf":"https://arxiv.org/pdf/1911.11033v4.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":"gating-revisited-deep-multi-layer-rnns-that-1","repo_url":"https://github.com/0zgur0/STAR_Network","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"gating-revisited-deep-multi-layer-rnns-that-1","repo_url":"https://github.com/croros/STAR_Network_Pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"gating-revisited-deep-multi-layer-rnns-that-1","repo_url":"https://github.com/titu1994/tf_star_rnn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"action-recognition-in-videos-2","task_name":"Action Recognition In Videos"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"music-modeling","task_name":"Music Modeling"},{"task_slug":"sequential-image-classification","task_name":"Sequential Image Classification"}],"methods":[{"method_slug":"gru","method_name":"GRU"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-recognition-in-videos-on-jester-1","task":"Action Recognition In Videos","dataset":"Jester (Gesture Recognition)","model":"convSTAR","rank_in_archive_order":6,"of":9,"metrics":{"Val":"92.7"},"uses_additional_data":false},{"leaderboard":"/sota/language-modelling-on-penn-treebank-character","task":"Language Modelling","dataset":"Penn Treebank (Character Level)","model":"STAR","rank_in_archive_order":17,"of":20,"metrics":{"Bit per Character (BPC)":"1.30"},"uses_additional_data":false},{"leaderboard":"/sota/sequential-image-classification-on-sequential","task":"Sequential Image Classification","dataset":"Sequential MNIST","model":"STAR","rank_in_archive_order":29,"of":30,"metrics":{"Unpermuted Accuracy":"99.4%"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}