Fu-Rec:Multi-Task Learning Recommendation Model Fusing Neighbor-Discrimination and Self-Discrimination

作     者：ZHENG Sirui HUANG Bo LIU Jin ZENG Guohui YIN Ling LI Zhi SUN Tie ZHENG Sirui;HUANG Bo;LIU Jin;ZENG Guohui;YIN Ling;LI Zhi;SUN Tie

作者机构：School of Electronic and Electrical EngineeringShanghai University of Engineering ScienceShanghai 201600China School of ComputerWuhan UniversityWuhan 430072HubeiChina School of Computer Science and EngineeringGuangxi Normal UniversityGuilin 541004GuangxiChina AIoT Manufacturing Solutions Technology Co.Ltd.Hefei 230000AnhuiChina 

出 版 物：《Wuhan University Journal of Natural Sciences》 (武汉大学学报（自然科学英文版）)

年 卷 期：2024年第29卷第2期

页      面：134-144页

核心收录：

学科分类：081203[工学-计算机应用技术] 08[工学] 0835[工学-软件工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Supported by the Scientific and Technological Innovation 2030-Major Project of New Generation Artificial Intelligence(2020AAA0109300) Science and Technology Commission of Shanghai Municipality(21DZ2203100) 2023 Anhui Province Key Research and Development Plan Project-Special Project of Science and Technology Cooperation(2023i11020002) 

主　　题：self-supervised learning recommendation system contrastive learning multi-task learning 

摘      要：In recent years,self-supervised learning has achieved great success in areas such as computer vision and natural language processing because it can mine supervised signals from unlabeled data and reduce the reliance on manual ***,the currently generated self-supervised signals are either neighbor discrimination or self-discrimination,and there is no model to integrate neighbor discrimination and *** on this,this paper proposes Fu-Rec that integrates neighbor-discrimination contrastive learning and self-discrimination contrastive learning,which consists of three modules:(1)neighbor-discrimination contrastive learning,(2)selfdiscrimination contrastive learning,and(3)recommendation *** neighbor-discrimination contrastive learning and selfdiscrimination contrastive learning tasks are used as auxiliary tasks to assist the recommendation *** Fu-Rec model effectively utilizes the respective advantages of neighbor-discrimination and self-discrimination to consider the information of the user’s neighbors as well as the user and the item itself for the recommendation,which results in better performance of the recommendation *** results on several public datasets demonstrate the effectiveness of the Fu-Rec proposed in this paper.