Towards Universal Sequence Representation Learning for Recommender Systems

被引:118
作者
Hou, Yupeng [1 ]
Mu, Shanlei [2 ]
Zhao, Wayne Xin [1 ,4 ]
Li, Yaliang [3 ]
Ding, Bolin [3 ]
Wen, Ji-Rong [1 ]
机构
[1] Renmin Univ China, Gaoling Sch Artificial Intelligence, Beijing Key Lab Big Data Management & Anal Method, Beijing, Peoples R China
[2] Renmin Univ China, Sch Informat, Beijing, Peoples R China
[3] Alibaba Grp, Hangzhou, Peoples R China
[4] Beijing Acad Artificial Intelligence, Beijing 100084, Peoples R China
来源
PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, KDD 2022 | 2022年
基金
中国国家自然科学基金; 北京市自然科学基金;
关键词
Sequential Recommendation; Universal Representation Learning;
D O I
10.1145/3534678.3539381
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In order to develop effective sequential recommenders, a series of sequence representation learning (SRL) methods are proposed to model historical user behaviors. Most existing SRL methods rely on explicit item IDs for developing the sequence models to better capture user preference. Though effective to some extent, these methods are difficult to be transferred to new recommendation scenarios, due to the limitation by explicitly modeling item IDs. To tackle this issue, we present a novel universal sequence representation learning approach, named UniSRec. The proposed approach utilizes the associated description text of items to learn transferable representations across different recommendation scenarios. For learning universal item representations, we design a lightweight item encoding architecture based on parametric whitening and mixture-of-experts enhanced adaptor. For learning universal sequence representations, we introduce two contrastive pre-training tasks by sampling multi-domain negatives. With the pre-trained universal sequence representation model, our approach can be effectively transferred to new recommendation domains or platforms in a parameter-efficient way, under either inductive or transductive settings. Extensive experiments conducted on real-world datasets demonstrate the effectiveness of the proposed approach. Especially, our approach also leads to a performance improvement in a cross-platform setting, showing the strong transferability of the proposed universal SRL method. The code and pre-trained model are available at: https://github.com/RUCAIBox/UniSRec.
引用
收藏
页码:585 / 593
页数:9
相关论文
共 39 条
[1]  
[Anonymous], 2012, SIGKDD
[2]   Representation Learning: A Review and New Perspectives [J].
Bengio, Yoshua ;
Courville, Aaron ;
Vincent, Pascal .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2013, 35 (08) :1798-1828
[3]   Sequential Recommendation with Graph Neural Networks [J].
Chang, Jianxin ;
Gao, Chen ;
Zheng, Yu ;
Hui, Yiqun ;
Niu, Yanan ;
Song, Yang ;
Jin, Depeng ;
Li, Yong .
SIGIR '21 - PROCEEDINGS OF THE 44TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL, 2021, :378-387
[4]  
Devlin J, 2019, 2019 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL HLT 2019), VOL. 1, P4171
[5]  
Ding H., 2021, arXiv
[6]   Locker: Locally Constrained Self-Attentive Sequential Recommendation [J].
He, Zhankui ;
Zhao, Handong ;
Wang, Zhaowen ;
Lin, Zhe ;
Kale, Ajinkya ;
McAuley, Julian .
PROCEEDINGS OF THE 30TH ACM INTERNATIONAL CONFERENCE ON INFORMATION & KNOWLEDGE MANAGEMENT, CIKM 2021, 2021, :3088-3092
[7]   Recurrent Neural Networks with Top-k Gains for Session-based Recommendations [J].
Hidasi, Balazs ;
Karatzoglou, Alexandros .
CIKM'18: PROCEEDINGS OF THE 27TH ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, 2018, :843-852
[8]   CORE: Simple and Effective Session-based Recommendation within Consistent Representation Space [J].
Hou, Yupeng ;
Hu, Binbin ;
Zhang, Zhiqiang ;
Zhao, Wayne Xin .
PROCEEDINGS OF THE 45TH INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL (SIGIR '22), 2022, :1796-1801
[9]   CoNet: Collaborative Cross Networks for Cross-Domain Recommendation [J].
Hu, Guangneng ;
Zhang, Yu ;
Yang, Qiang .
CIKM'18: PROCEEDINGS OF THE 27TH ACM INTERNATIONAL CONFERENCE ON INFORMATION AND KNOWLEDGE MANAGEMENT, 2018, :667-676
[10]  
Huang Junjie, 2021, Findings of EMNLP