Causal Embeddings for Recommendation

被引:170
|
作者
Bonner, Stephen [1 ,2 ]
Vasile, Flavian [1 ]
机构
[1] Criteo AI Labs, Paris, France
[2] Univ Durham, Dept Comp Sci, Durham, England
关键词
Recommender Systems; Causality; Embeddings; Neural Networks; Counterfactual Inference;
D O I
10.1145/3240323.3240360
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Many current applications use recommendations in order to modify the natural user behavior, such as to increase the number of sales or the time spent on a website. This results in a gap between the final recommendation objective and the classical setup where recommendation candidates are evaluated by their coherence with past user behavior, by predicting either the missing entries in the user-item matrix, or the most likely next event. To bridge this gap, we optimize a recommendation policy for the task of increasing the desired outcome versus the organic user behavior. We show this is equivalent to learning to predict recommendation outcomes under a fully random recommendation policy. To this end, we propose a new domain adaptation algorithm that learns from logged data containing outcomes from a biased recommendation policy and predicts recommendation outcomes according to random exposure. We compare our method against state-of-the-art factorization methods, in addition to new approaches of causal recommendation and show significant improvements.
引用
收藏
页码:104 / 112
页数:9
相关论文
共 50 条
  • [41] Unifying paragraph embeddings and neural collaborative filtering for hybrid recommendation
    Zhang, Yihao
    Liu, Zhi
    Sang, Chunyan
    APPLIED SOFT COMPUTING, 2021, 106
  • [42] Perceiving the Next Choice with Comprehensive Transaction Embeddings for Online Recommendation
    Wang, Shoujin
    Hu, Liang
    Cao, Longbing
    MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2017, PT II, 2017, 10535 : 285 - 302
  • [43] A Survey on Debiasing Recommendation Based on Causal Inference
    Yang, Xin-Xin
    Liu, Zhen
    Lu, Si-Bo
    Yuan, Ya-Fan
    Sun, Yong-Qi
    Jisuanji Xuebao/Chinese Journal of Computers, 2024, 47 (10): : 2307 - 2332
  • [44] Learning Knowledge Embeddings with Prior Weights for Sparse Interaction Recommendation
    Yang, Deqing
    Guo, Zikai
    Xiao, Yanghua
    11TH IEEE INTERNATIONAL CONFERENCE ON KNOWLEDGE GRAPH (ICKG 2020), 2020, : 211 - 218
  • [45] Mitigating Hidden Confounding Effects for Causal Recommendation
    Zhu, Xinyuan
    Zhang, Yang
    Feng, Fuli
    Yang, Xun
    Wang, Dingxian
    He, Xiangnan
    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2024, 36 (09) : 4794 - 4805
  • [46] Revisiting Drug Recommendation From a Causal Perspective
    Zhang, Junjie
    Zang, Xuan
    Chen, Hao
    Yan, Xiaowei
    Tang, Buzhou
    IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2025, 29 (02) : 1525 - 1533
  • [47] Cross-domain Recommendation with Bridge-Item Embeddings
    Gao, Chen
    Li, Yong
    Feng, Fuli
    Chen, Xiangning
    Zhao, Kai
    He, Xiangnan
    Jin, Depeng
    ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA, 2022, 16 (01)
  • [48] Learning Student and Content Embeddings for Personalized Lesson Sequence Recommendation
    Reddy, Siddharth
    Labutov, Igor
    Joachims, Thorsten
    PROCEEDINGS OF THE THIRD (2016) ACM CONFERENCE ON LEARNING @ SCALE (L@S 2016), 2016, : 93 - 96
  • [49] A Place Recommendation Approach Using Word Embeddings in Conceptual Spaces
    Abbasi, Omid R. R.
    Alesheikh, Ali A. A.
    IEEE ACCESS, 2023, 11 : 11871 - 11879
  • [50] Hypergraph User Embeddings and Session Contrastive Learning for POI Recommendation
    Zhang, Yan
    Wang, Bin
    Zhang, Qian
    Zhu, Sulei
    Ma, Yan
    IEEE ACCESS, 2025, 13 : 17983 - 17995