{"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/factorization-machines","title":"Factorization Machines","arxiv_id":null,"date":"2011-01-20","proceeding":"2010/01/01 2010 1","authors":["Steffen Rendle"],"abstract":"In this paper, we introduce Factorization Machines\r\n(FM) which are a new model class that combines the advantages\r\nof Support Vector Machines (SVM) with factorization models.\r\nLike SVMs, FMs are a general predictor working with any\r\nreal valued feature vector. In contrast to SVMs, FMs model all\r\ninteractions between variables using factorized parameters. Thus\r\nthey are able to estimate interactions even in problems with huge\r\nsparsity (like recommender systems) where SVMs fail. We show\r\nthat the model equation of FMs can be calculated in linear time\r\nand thus FMs can be optimized directly. So unlike nonlinear\r\nSVMs, a transformation in the dual form is not necessary and\r\nthe model parameters can be estimated directly without the need\r\nof any support vector in the solution. We show the relationship\r\nto SVMs and the advantages of FMs for parameter estimation\r\nin sparse settings.\r\nOn the other hand there are many different factorization models like matrix factorization, parallel factor analysis or specialized\r\nmodels like SVD++, PITF or FPMC. The drawback of these\r\nmodels is that they are not applicable for general prediction tasks\r\nbut work only with special input data. Furthermore their model\r\nequations and optimization algorithms are derived individually\r\nfor each task. We show that FMs can mimic these models just\r\nby specifying the input data (i.e. the feature vectors). This makes\r\nFMs easily applicable even for users without expert knowledge\r\nin factorization models.\r\nIndex Terms—factorization machine; sparse data; tensor factorization; support vector machine","url_abs":"https://ieeexplore.ieee.org/document/5694074","url_pdf":"https://www.csie.ntu.edu.tw/~b97053/paper/Rendle2010FM.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":"factorization-machines","repo_url":"https://github.com/PaddlePaddle/PaddleRec/tree/release/2.1.0/models/rank/fm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null},{"paper_slug":"factorization-machines","repo_url":"https://github.com/UlionTse/mlgb","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"},{"task_slug":"parameter-estimation","task_name":"parameter estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}