{"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/conet-collaborative-cross-networks-for-cross","title":"CoNet: Collaborative Cross Networks for Cross-Domain Recommendation","arxiv_id":"1804.06769","date":"2018-04-18","proceeding":null,"authors":["Guang-Neng Hu","Yu Zhang","Qiang Yang"],"abstract":"The cross-domain recommendation technique is an effective way of alleviating\nthe data sparse issue in recommender systems by leveraging the knowledge from\nrelevant domains. Transfer learning is a class of algorithms underlying these\ntechniques. In this paper, we propose a novel transfer learning approach for\ncross-domain recommendation by using neural networks as the base model. In\ncontrast to the matrix factorization based cross-domain techniques, our method\nis deep transfer learning, which can learn complex user-item interaction\nrelationships. We assume that hidden layers in two base networks are connected\nby cross mappings, leading to the collaborative cross networks (CoNet). CoNet\nenables dual knowledge transfer across domains by introducing cross connections\nfrom one base network to another and vice versa. CoNet is achieved in\nmulti-layer feedforward networks by adding dual connections and joint loss\nfunctions, which can be trained efficiently by back-propagation. The proposed\nmodel is thoroughly evaluated on two large real-world datasets. It outperforms\nbaselines by relative improvements of 7.84\\% in NDCG. We demonstrate the\nnecessity of adaptively selecting representations to transfer. Our model can\nreduce tens of thousands training examples comparing with non-transfer methods\nand still has the competitive performance with them.","url_abs":"http://arxiv.org/abs/1804.06769v3","url_pdf":"http://arxiv.org/pdf/1804.06769v3.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":"conet-collaborative-cross-networks-for-cross","repo_url":"https://github.com/njuhugn/CoNet","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"recommendation-systems","task_name":"Recommendation Systems"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.06769","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}