{"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/sequential-variational-autoencoders-for","title":"Sequential Variational Autoencoders for Collaborative Filtering","arxiv_id":"1811.09975","date":"2018-11-25","proceeding":null,"authors":["Noveen Sachdeva","Giuseppe Manco","Ettore Ritacco","Vikram Pudi"],"abstract":"Variational autoencoders were proven successful in domains such as computer\nvision and speech processing. Their adoption for modeling user preferences is\nstill unexplored, although recently it is starting to gain attention in the\ncurrent literature. In this work, we propose a model which extends variational\nautoencoders by exploiting the rich information present in the past preference\nhistory. We introduce a recurrent version of the VAE, where instead of passing\na subset of the whole history regardless of temporal dependencies, we rather\npass the consumption sequence subset through a recurrent neural network. At\neach time-step of the RNN, the sequence is fed through a series of\nfully-connected layers, the output of which models the probability distribution\nof the most likely future preferences. We show that handling temporal\ninformation is crucial for improving the accuracy of the VAE: In fact, our\nmodel beats the current state-of-the-art by valuable margins because of its\nability to capture temporal dependencies among the user-consumption sequence\nusing the recurrent encoder still keeping the fundamentals of variational\nautoencoders intact.","url_abs":"http://arxiv.org/abs/1811.09975v1","url_pdf":"http://arxiv.org/pdf/1811.09975v1.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":"sequential-variational-autoencoders-for","repo_url":"https://github.com/noveens/svae_cf","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/collaborative-filtering-on-movielens-1m","task":"Recommendation Systems","dataset":"MovieLens 1M","model":"SVAE","rank_in_archive_order":27,"of":31,"metrics":{"nDCG@100":"0.2993"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.09975","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}