{"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/temporal-attention-gated-model-for-robust","title":"Temporal Attention-Gated Model for Robust Sequence Classification","arxiv_id":"1612.00385","date":"2016-12-01","proceeding":"CVPR 2017 7","authors":["Wenjie Pei","Tadas Baltrušaitis","David M. J. Tax","Louis-Philippe Morency"],"abstract":"Typical techniques for sequence classification are designed for\nwell-segmented sequences which have been edited to remove noisy or irrelevant\nparts. Therefore, such methods cannot be easily applied on noisy sequences\nexpected in real-world applications. In this paper, we present the Temporal\nAttention-Gated Model (TAGM) which integrates ideas from attention models and\ngated recurrent networks to better deal with noisy or unsegmented sequences.\nSpecifically, we extend the concept of attention model to measure the relevance\nof each observation (time step) of a sequence. We then use a novel gated\nrecurrent network to learn the hidden representation for the final prediction.\nAn important advantage of our approach is interpretability since the temporal\nattention weights provide a meaningful value for the salience of each time step\nin the sequence. We demonstrate the merits of our TAGM approach, both for\nprediction accuracy and interpretability, on three different tasks: spoken\ndigit recognition, text-based sentiment analysis and visual event recognition.","url_abs":"http://arxiv.org/abs/1612.00385v2","url_pdf":"http://arxiv.org/pdf/1612.00385v2.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":"temporal-attention-gated-model-for-robust","repo_url":"https://github.com/wenjiepei/TAGM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":null}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"model","task_name":"model"}],"methods":[{"method_slug":"interpretability","method_name":"Interpretability"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1612.00385","atlas_url":"https://app.syntology.ai/?focus=1612.00385","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}