Information Theoretic Counterfactual Learning from Missing-Not-At-Random Feedback

被引:0
作者
Wang, Zifeng [1 ]
Chen, Xi [2 ]
Wen, Rui [2 ]
Huang, Shao-Lun [1 ]
Kuruoglu, Ercan E. [1 ,3 ]
Zheng, Yefeng [2 ]
机构
[1] Tsinghua Univ, Tsinghua Berkeley Shenzhen Inst, Beijing, Peoples R China
[2] Tencent, Jarvis Lab, Shenzhen, Peoples R China
[3] CNR, Inst Sci & Technol Informat, Pisa, Italy
来源
ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 33, NEURIPS 2020 | 2020年 / 33卷
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Counterfactual learning for dealing with missing-not-at-random data (MNAR) is an intriguing topic in the recommendation literature since MNAR data are ubiquitous in modern recommender systems. Missing-at-random (MAR) data, namely randomized controlled trials (RCTs), are usually required by most previous counterfactual learning methods for debiasing learning. However, the execution of RCTs is extraordinarily expensive in practice. To circumvent the use of RCTs, we build an information-theoretic counterfactual variational information bottleneck (CVIB), as an alternative for debiasing learning without RCTs. By separating the task-aware mutual information term in the original information bottleneck Lagrangian into factual and counterfactual parts, we derive a contrastive information loss and an additional output confidence penalty, which facilitates balanced learning between the factual and counterfactual domains. Empirical evaluation on real-world datasets shows that our CVIB significantly enhances both shallow and deep models, which sheds light on counterfactual learning in recommendation that goes beyond RCTs.
引用
收藏
页数:11
相关论文
共 37 条
[1]  
Achille A, 2018, J MACH LEARN RES, V19
[2]   Information Dropout: Learning Optimal Representations Through Noisy Computation [J].
Achille, Alessandro ;
Soatto, Stefano .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2018, 40 (12) :2897-2905
[3]  
Alemi A. A., 2016, PROC INT C LEARN REP
[4]  
[Anonymous], 2008, ACM C KNOWL DISC DAT, DOI DOI 10.1145/1401890.1401944
[5]  
[Anonymous], 2016, INT C MACH LEARN
[6]  
Berger AL, 1996, COMPUT LINGUIST, V22, P39
[7]  
Bialek William, 2000, ACC
[8]   Causal Embeddings for Recommendation [J].
Bonner, Stephen ;
Vasile, Flavian .
12TH ACM CONFERENCE ON RECOMMENDER SYSTEMS (RECSYS), 2018, :104-112
[9]   A METHOD FOR ASSESSING THE QUALITY OF A RANDOMIZED CONTROL TRIAL [J].
CHALMERS, TC ;
SMITH, H ;
BLACKBURN, B ;
SILVERMAN, B ;
SCHROEDER, B ;
REITMAN, D ;
AMBROZ, A .
CONTROLLED CLINICAL TRIALS, 1981, 2 (01) :31-49
[10]  
Chen X, 2016, 30 C NEURAL INFORM P, V29