{"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/deep-collective-matrix-factorization-for","title":"Deep Collective Matrix Factorization for Augmented Multi-View Learning","arxiv_id":"1811.11427","date":"2018-11-28","proceeding":null,"authors":["Ragunathan Mariappan","Vaibhav Rajan"],"abstract":"Learning by integrating multiple heterogeneous data sources is a common\nrequirement in many tasks. Collective Matrix Factorization (CMF) is a technique\nto learn shared latent representations from arbitrary collections of matrices.\nIt can be used to simultaneously complete one or more matrices, for predicting\nthe unknown entries. Classical CMF methods assume linearity in the interaction\nof latent factors which can be restrictive and fails to capture complex\nnon-linear interactions. In this paper, we develop the first deep-learning\nbased method, called dCMF, for unsupervised learning of multiple shared\nrepresentations, that can model such non-linear interactions, from an arbitrary\ncollection of matrices. We address optimization challenges that arise due to\ndependencies between shared representations through Multi-Task Bayesian\nOptimization and design an acquisition function adapted for collective learning\nof hyperparameters. Our experiments show that dCMF significantly outperforms\nprevious CMF algorithms in integrating heterogeneous data for predictive\nmodeling. Further, on two tasks - recommendation and prediction of gene-disease\nassociation - dCMF outperforms state-of-the-art matrix completion algorithms\nthat can utilize auxiliary sources of information.","url_abs":"http://arxiv.org/abs/1811.11427v2","url_pdf":"http://arxiv.org/pdf/1811.11427v2.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":"deep-collective-matrix-factorization-for","repo_url":"https://bitbucket.org/cdal/dcmf","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"bayesian-optimization","task_name":"Bayesian Optimization"},{"task_slug":"multi-view-learning","task_name":"MULTI-VIEW LEARNING"},{"task_slug":"matrix-completion","task_name":"Matrix Completion"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}