{"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/predicting-pairwise-relations-with-neural","title":"Predicting Pairwise Relations with Neural Similarity Encoders","arxiv_id":"1702.01824","date":"2017-02-06","proceeding":null,"authors":["Franziska Horn","Klaus-Robert Müller"],"abstract":"Matrix factorization is at the heart of many machine learning algorithms, for\nexample, dimensionality reduction (e.g. kernel PCA) or recommender systems\nrelying on collaborative filtering. Understanding a singular value\ndecomposition (SVD) of a matrix as a neural network optimization problem\nenables us to decompose large matrices efficiently while dealing naturally with\nmissing values in the given matrix. But most importantly, it allows us to learn\nthe connection between data points' feature vectors and the matrix containing\ninformation about their pairwise relations. In this paper we introduce a novel\nneural network architecture termed Similarity Encoder (SimEc), which is\ndesigned to simultaneously factorize a given target matrix while also learning\nthe mapping to project the data points' feature vectors into a similarity\npreserving embedding space. This makes it possible to, for example, easily\ncompute out-of-sample solutions for new data points. Additionally, we\ndemonstrate that SimEc can preserve non-metric similarities and even predict\nmultiple pairwise relations between data points at once.","url_abs":"http://arxiv.org/abs/1702.01824v2","url_pdf":"http://arxiv.org/pdf/1702.01824v2.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":"predicting-pairwise-relations-with-neural","repo_url":"https://github.com/cod3licious/simec","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"},{"task_slug":"dimensionality-reduction","task_name":"Dimensionality Reduction"},{"task_slug":"missing-values","task_name":"Missing Values"},{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}