{"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/memory-fusion-network-for-multi-view","title":"Memory Fusion Network for Multi-view Sequential Learning","arxiv_id":"1802.00927","date":"2018-02-03","proceeding":null,"authors":["Amir Zadeh","Paul Pu Liang","Navonil Mazumder","Soujanya Poria","Erik Cambria","Louis-Philippe Morency"],"abstract":"Multi-view sequential learning is a fundamental problem in machine learning\ndealing with multi-view sequences. In a multi-view sequence, there exists two\nforms of interactions between different views: view-specific interactions and\ncross-view interactions. In this paper, we present a new neural architecture\nfor multi-view sequential learning called the Memory Fusion Network (MFN) that\nexplicitly accounts for both interactions in a neural architecture and\ncontinuously models them through time. The first component of the MFN is called\nthe System of LSTMs, where view-specific interactions are learned in isolation\nthrough assigning an LSTM function to each view. The cross-view interactions\nare then identified using a special attention mechanism called the Delta-memory\nAttention Network (DMAN) and summarized through time with a Multi-view Gated\nMemory. Through extensive experimentation, MFN is compared to various proposed\napproaches for multi-view sequential learning on multiple publicly available\nbenchmark datasets. MFN outperforms all the existing multi-view approaches.\nFurthermore, MFN outperforms all current state-of-the-art models, setting new\nstate-of-the-art results for these multi-view datasets.","url_abs":"http://arxiv.org/abs/1802.00927v1","url_pdf":"http://arxiv.org/pdf/1802.00927v1.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":"memory-fusion-network-for-multi-view","repo_url":"https://github.com/clin366/MFN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"memory-fusion-network-for-multi-view","repo_url":"https://github.com/pliang279/MFN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"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":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.00927","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}