{"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/collaborative-memory-network-for","title":"Collaborative Memory Network for Recommendation Systems","arxiv_id":"1804.10862","date":"2018-04-29","proceeding":null,"authors":["Travis Ebesu","Bin Shen","Yi Fang"],"abstract":"Recommendation systems play a vital role to keep users engaged with\npersonalized content in modern online platforms. Deep learning has\nrevolutionized many research fields and there is a recent surge of interest in\napplying it to collaborative filtering (CF). However, existing methods compose\ndeep learning architectures with the latent factor model ignoring a major class\nof CF models, neighborhood or memory-based approaches. We propose Collaborative\nMemory Networks (CMN), a deep architecture to unify the two classes of CF\nmodels capitalizing on the strengths of the global structure of latent factor\nmodel and local neighborhood-based structure in a nonlinear fashion. Motivated\nby the success of Memory Networks, we fuse a memory component and neural\nattention mechanism as the neighborhood component. The associative addressing\nscheme with the user and item memories in the memory module encodes complex\nuser-item relations coupled with the neural attention mechanism to learn a\nuser-item specific neighborhood. Finally, the output module jointly exploits\nthe neighborhood with the user and item memories to produce the ranking score.\nStacking multiple memory modules together yield deeper architectures capturing\nincreasingly complex user-item relations. Furthermore, we show strong\nconnections between CMN components, memory networks and the three classes of CF\nmodels. Comprehensive experimental results demonstrate the effectiveness of CMN\non three public datasets outperforming competitive baselines. Qualitative\nvisualization of the attention weights provide insight into the model's\nrecommendation process and suggest the presence of higher order interactions.","url_abs":"http://arxiv.org/abs/1804.10862v1","url_pdf":"http://arxiv.org/pdf/1804.10862v1.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":"collaborative-memory-network-for","repo_url":"https://github.com/tebesu/CollaborativeMemoryNetwork","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"collaborative-memory-network-for","repo_url":"https://github.com/ArgentLo/PPNW-KAIS","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"collaborative-memory-network-for","repo_url":"https://github.com/IamAdiSri/cmn4recosys","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"collaborative-memory-network-for","repo_url":"https://github.com/caroprese/CollaborativeMemoryNetwork-master","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"collaborative-memory-network-for","repo_url":"https://github.com/rouxiaoxiao/CMNPlus","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"},{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.10862","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}