research-article

Public Access

Reconsidering Learning Objectives in Unbiased Recommendation: A Distribution Shift Perspective

Authors:

Suhang WangAuthors Info & Claims

KDD '23: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

Pages 2764 - 2775

https://doi.org/10.1145/3580305.3599487

Published: 04 August 2023 Publication History

Abstract

This work studies the problem of learning unbiased algorithms from biased feedback for recommendation. We address this problem from a novel distribution shift perspective. Recent works in unbiased recommendation have advanced the state-of-the-art with various techniques such as re-weighting, multi-task learning, and meta-learning. Despite their empirical successes, most of them lack theoretical guarantees, forming non-negligible gaps between theories and recent algorithms. In this paper, we propose a theoretical understanding of why existing unbiased learning objectives work for unbiased recommendation. We establish a close connection between unbiased recommendation and distribution shift, which shows that existing unbiased learning objectives implicitly align biased training and unbiased test distributions. Built upon this connection, we develop two generalization bounds for existing unbiased learning methods and analyze their learning behavior. Besides, as a result of the distribution shift, we further propose a principled framework, Adversarial Self-Training (AST), for unbiased recommendation. Extensive experiments on real-world and semi-synthetic datasets demonstrate the effectiveness of AST.

Supplementary Material

MP4 File (GMT20230625-204521_Recording_1600x900.mp4)

We study the problem of learning unbiased algorithms from biased feedback for recommendation and address this problem from a novel distribution shift perspective.

Download
2.91 MB

References

[1]

Heejung Bang and James M Robins. 2005. Doubly robust estimation in missing data and causal inference models. Biometrics.

[2]

Shai Ben-David, John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman Vaughan. 2010. A theory of learning from different domains. Machine learning.

[3]

Shai Ben-David, John Blitzer, Koby Crammer, Fernando Pereira, et al. 2007. Analysis of representations for domain adaptation. NIPS.

[4]

David Berthelot, Nicholas Carlini, Ian Goodfellow, Nicolas Papernot, Avital Oliver, and Colin A Raffel. 2019. MixMatch: A Holistic Approach to Semi-Supervised Learning. NIPS.

[5]

John Blitzer, Koby Crammer, Alex Kulesza, Fernando Pereira, and Jennifer Wortman. 2007. Learning Bounds for Domain Adaptation. NIPS.

[6]

Stephen Bonner and Flavian Vasile. 2018. Causal embeddings for recommendation. RecSys.

[7]

Jiawei Chen, Hande Dong, Yang Qiu, Xiangnan He, Xin Xin, Liang Chen, Guli Lin, and Keping Yang. 2021. AutoDebias: Learning to Debias for Recommendation. SIGIR.

[8]

Yining Chen, Colin Wei, Ananya Kumar, and Tengyu Ma. 2020a. Self-training avoids using spurious features under domain shift. NIPS (2020), 21061--21071.

[9]

Zhihong Chen, Rong Xiao, Chenliang Li, Gangfeng Ye, Haochuan Sun, and Hongbo Deng. 2020b. Esam: Discriminative domain adaptation with non-displayed items to improve long-tail performance. SIGIR.

Digital Library

[10]

Imre Csiszár and János Körner. 2011. Information theory: coding theorems for discrete memoryless systems.

[11]

Sihao Ding, Peng Wu, Fuli Feng, Yitong Wang, Xiangnan He, Yong Liao, and Yongdong Zhang. 2022. Addressing Unmeasured Confounder for Recommendation with Sensitivity Analysis. In KDD.

[12]

Miroslav Dudík, John Langford, and Lihong Li. 2011. Doubly robust policy evaluation and learning. ICML.

[13]

Yaroslav Ganin and Victor Lempitsky. 2015. Unsupervised domain adaptation by backpropagation. ICML.

[14]

Yves Grandvalet and Yoshua Bengio. 2004. Semi-supervised learning by entropy minimization. NIPS.

[15]

Siyuan Guo, Lixin Zou, Yiding Liu, Wenwen Ye, Suqi Cheng, Shuaiqiang Wang, Hechang Chen, Dawei Yin, and Yi Chang. 2021. Enhanced Doubly Robust Learning for Debiasing Post-Click Conversion Rate Estimation. SIGIR.

[16]

Zhimeng Guo, Teng Xiao, Charu Aggarwal, Hui Liu, and Suhang Wang. 2023. Counterfactual Learning on Graphs: A Survey. arXiv preprint arXiv:2304.01391 (2023).

[17]

Shantanu Gupta, Hao Wang, Zachary Lipton, and Yuyang Wang. 2021. Correcting Exposure Bias for Link Recommendation. ICML.

[18]

Xiangnan He, Zhankui He, Xiaoyu Du, and Tat-Seng Chua. 2018. Adversarial personalized ranking for recommendation. In SIGIR.

[19]

Xiangnan He, Lizi Liao, Hanwang Zhang, Liqiang Nie, Xia Hu, and Tat-Seng Chua. 2017. Neural collaborative filtering. WWW.

[20]

Fredrik D Johansson, David Sontag, and Rajesh Ranganath. 2019. Support and invertibility in domain-invariant representations. In AISTATS.

[21]

Yehuda Koren, Robert Bell, and Chris Volinsky. 2009. Matrix factorization techniques for recommender systems. Computer.

[22]

Adit Krishnan, Ashish Sharma, Aravind Sankar, and Hari Sundaram. 2018. An adversarial approach to improve long-tail performance in neural collaborative filtering. CIKM.

[23]

Sanjay Krishnan, Jay Patel, Michael J Franklin, and Ken Goldberg. 2014. A methodology for learning, analyzing, and mitigating social influence bias in recommender systems. RecSys.

[24]

Jingjing Li, Erpeng Chen, Zhengming Ding, Lei Zhu, Ke Lu, and Heng Tao Shen. 2020. Maximum density divergence for domain adaptation. PAMI.

[25]

Zinan Lin, Dugang Liu, Weike Pan, and Zhong Ming. 2021. Transfer Learning in Collaborative Recommendation for Bias Reduction. RecSys.

[26]

Dugang Liu, Pengxiang Cheng, Zhenhua Dong, Xiuqiang He, Weike Pan, and Zhong Ming. 2020. A general knowledge distillation framework for counterfactual recommendation via uniform data. SIGIR.

[27]

Dugang Liu, Pengxiang Cheng, Zinan Lin, Jinwei Luo, Zhenhua Dong, Xiuqiang He, Weike Pan, and Zhong Ming. 2022. KDCRec: Knowledge Distillation for Counterfactual Recommendation Via Uniform Data. TKDE (2022).

[28]

Dugang Liu, Pengxiang Cheng, Hong Zhu, Zhenhua Dong, Xiuqiang He, Weike Pan, and Zhong Ming. 2021a. Mitigating Confounding Bias in Recommendation via Information Bottleneck. RecSys.

[29]

Weiming Liu, Jiajie Su, Chaochao Chen, and Xiaolin Zheng. 2021b. Leveraging Distribution Alignment via Stein Path for Cross-Domain Cold-Start Recommendation. NeurIPS.

[30]

Mingsheng Long, Zhangjie Cao, Jianmin Wang, and Michael I Jordan. 2018. Conditional adversarial domain adaptation. NIPS.

[31]

Benjamin M Marlin, Richard S Zemel, Sam Roweis, and Malcolm Slaney. 2007. Collaborative filtering and the missing at random assumption. UAI.

[32]

Sebastian Nowozin, Botond Cseke, and Ryota Tomioka. 2016. f-gan: Training generative neural samplers using variational divergence minimization. NIPS.

[33]

Paul R Rosenbaum and Donald B Rubin. 1983. The central role of the propensity score in observational studies for causal effects. Biometrika.

[34]

Donald B Rubin. 1974. Estimating causal effects of treatments in randomized and nonrandomized studies. journal of educational Psychology.

[35]

Noveen Sachdeva, Yi Su, and Thorsten Joachims. 2020. Off-policy bandits with deficient support. KDD.

[36]

Yuta Saito. 2020. Asymmetric Tri-training for Debiasing Missing-Not-At-Random Explicit Feedback. SIGIR.

[37]

Yuta Saito, Suguru Yaginuma, Yuta Nishino, Hayato Sakata, and Kazuhide Nakata. 2020. Unbiased recommender learning from missing-not-at-random implicit feedback. WSDM.

[38]

Tobias Schnabel, Adith Swaminathan, Ashudeep Singh, Navin Chandak, and Thorsten Joachims. 2016. Recommendations as treatments: Debiasing learning and evaluation. ICML.

[39]

Ashish Shrivastava, Tomas Pfister, Oncel Tuzel, Joshua Susskind, Wenda Wang, and Russell Webb. 2017. Learning from simulated and unsupervised images through adversarial training. CVPR.

[40]

Jun Shu, Qi Xie, Lixuan Yi, Qian Zhao, Sanping Zhou, Zongben Xu, and Deyu Meng. 2019. Meta-weight-net: Learning an explicit mapping for sample weighting. NIPS.

[41]

Yi Su, Maria Dimakopoulou, Akshay Krishnamurthy, and Miroslav Dudík. 2020. Doubly robust off-policy evaluation with shrinkage. ICML.

[42]

Adith Swaminathan and Thorsten Joachims. 2015. The self-normalized estimator for counterfactual learning. NIPS.

[43]

James Victor Uspensky. 1937. Introduction to mathematical probability.

[44]

Vladimir N Vapnik. 1999. An overview of statistical learning theory. IEEE transactions on neural networks.

[45]

Mengting Wan and Julian McAuley. 2018. Item recommendation on monotonic behavior chains. In RecSys.

[46]

Qi Wan, Xiangnan He, Xiang Wang, Jiancan Wu, Wei Guo, and Ruiming Tang. 2022. Cross Pairwise Ranking for Unbiased Item Recommendation. In WWW.

[47]

Jun Wang, Lantao Yu, Weinan Zhang, Yu Gong, Yinghui Xu, Benyou Wang, Peng Zhang, and Dell Zhang. 2017. Irgan: A minimax game for unifying generative and discriminative information retrieval models. In SIGIR.

Digital Library

[48]

Xiaojie Wang, Rui Zhang, Yu Sun, and Jianzhong Qi. 2019. Doubly robust joint learning for recommendation on data missing not at random. ICML.

[49]

Xiaojie Wang, Rui Zhang, Yu Sun, and Jianzhong Qi. 2021. Combating Selection Biases in Recommender Systems with a Few Unbiased Ratings. WSDM.

[50]

Zifeng Wang, Xi Chen, Rui Wen, Shao-Lun Huang, Ercan Kuruoglu, and Yefeng Zheng. 2020. Information Theoretic Counterfactual Learning from Missing-Not-At-Random Feedback. NIPS.

[51]

Colin Wei, Kendrick Shen, Yining Chen, and Tengyu Ma. 2020. Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data. In ICLR.

[52]

Chenwang Wu, Defu Lian, Yong Ge, Zhihao Zhu, Enhong Chen, and Senchao Yuan. 2021a. Fight fire with fire: towards robust recommender systems via adversarial poisoning training. In SIGIR. 1074--1083.

[53]

Chuhan Wu, Fangzhao Wu, Xiting Wang, Yongfeng Huang, and Xing Xie. 2021b. Fairness-aware news recommendation with decomposed adversarial learning. In AAAI. 4462--4469.

[54]

Teng Xiao, Zhengyu Chen, Donglin Wang, and Suhang Wang. 2021. Learning how to propagate messages in graph neural networks. In KDD. 1894--1903.

[55]

Teng Xiao, Zhengyu Chen, and Suhang Wang. 2022. Representation Matters When Learning From Biased Feedback in Recommendation. In CIKM. 2220--2229.

[56]

Teng Xiao, Shangsong Liang, and Zaiqiao Meng. 2019. Hierarchical neural variational model for personalized sequential recommendation. In WWW.

[57]

Teng Xiao and Donglin Wang. 2021. A general offline reinforcement learning framework for interactive recommendation. In AAAI.

[58]

Teng Xiao and Suhang Wang. 2022. Towards unbiased and robust causal ranking for recommender systems. In WSDM. 1158--1167.

[59]

Da Xu, Chuanwei Ruan, Evren Korpeoglu, Sushant Kumar, and Kannan Achan. 2020. Adversarial Counterfactual Learning and Evaluation for Recommender System. NIPS.

[60]

Da Xu, Chuanwei Ruan, Evren Korpeoglu, Sushant Kumar, and Kannan Achan. 2021. Rethinking Neural vs. Matrix-Factorization Collaborative Filtering: the Theoretical Perspectives. ICML.

[61]

Longqi Yang, Yin Cui, Yuan Xuan, Chenyang Wang, Serge Belongie, and Deborah Estrin. 2018. Unbiased offline recommender evaluation for missing-not-at-random implicit feedback. RecSys.

[62]

Feng Yuan, Lina Yao, and Boualem Benatallah. 2019. DARec: Deep domain adaptation for cross-domain recommendation via transferring rating patterns. arXiv.

[63]

Han Zhao, Remi Tachet Des Combes, Kun Zhang, and Geoffrey Gordon. 2019. On learning invariant representations for domain adaptation. ICML.

[64]

Ziwei Zhu, Yun He, Yin Zhang, and James Caverlee. 2020. Unbiased Implicit Recommendation and Propensity Estimation via Combinational Joint Learning. RecSys.

Cited By

Zhang ZWang XChen HLi HZhu W(2024)Disentangled Dynamic Graph Attention Network for Out-of-Distribution Sequential RecommendationACM Transactions on Information Systems10.1145/370198843:1(1-42)Online publication date: 29-Oct-2024
https://dl.acm.org/doi/10.1145/3701988
Lv ZHe SZhan TZhang SZhang WChen JZhao ZWu FCai JKankanhalli MPrabhakaran BBoll SSubramanian RZheng LSingh VCesar PXie LXu D(2024)Semantic Codebook Learning for Dynamic Recommendation ModelsProceedings of the 32nd ACM International Conference on Multimedia10.1145/3664647.3680574(9611-9620)Online publication date: 28-Oct-2024
https://dl.acm.org/doi/10.1145/3664647.3680574
Chen ZXiao TWang DZhang M(2024)Pareto Graph Self-Supervised LearningICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)10.1109/ICASSP48485.2024.10447557(6630-6634)Online publication date: 14-Apr-2024
https://doi.org/10.1109/ICASSP48485.2024.10447557
Show More Cited By

Index Terms

Reconsidering Learning Objectives in Unbiased Recommendation: A Distribution Shift Perspective
1. Information systems
  1. Information retrieval

Recommendations

Toward Pareto Efficient Fairness-Utility Trade-off in Recommendation through Reinforcement Learning
WSDM '22: Proceedings of the Fifteenth ACM International Conference on Web Search and Data Mining

The issue of fairness in recommendation is becoming increasingly essential as Recommender Systems (RS) touch and influence more and more people in their daily lives. In fairness-aware recommendation, most of the existing algorithmic approaches mainly ...
Unbiased Learning for the Causal Effect of Recommendation
RecSys '20: Proceedings of the 14th ACM Conference on Recommender Systems

Increasing users’ positive interactions, such as purchases or clicks, is an important objective of recommender systems. Recommenders typically aim to select items that users will interact with. If the recommended items are purchased, an increase in ...
TCCM: Time and Content-Aware Causal Model for Unbiased News Recommendation
CIKM '23: Proceedings of the 32nd ACM International Conference on Information and Knowledge Management

Popularity bias significantly impacts news recommendation systems, as popular news articles receive more exposure and are often delivered to irrelevant users, resulting in unsatisfactory performance. Existing methods have not adequately addressed the ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

KDD '23: Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 2023

5996 pages

ISBN:9798400701030

DOI:10.1145/3580305

General Chairs:
Ambuj Singh
UC Santa Barbara, USA
,
Yizhou Sun
UC Los Angeles, USA
,
Program Chairs:
Leman Akoglu
Carnegie Mellon University, USA
,
Dimitrios Gunopulos
University of Athens, Greece
,
Xifeng Yan
UC Santa Barbara, USA
,
Ravi Kumar
Google, USA
,
Fatma Ozcan
Google, USA
,
Jieping Ye
Alibaba DAMO Academy

Copyright © 2023 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected].

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 04 August 2023

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Funding Sources

Cisco Faculty Research Award
Army Research Office (ARO)
National Science Foundation

Conference

KDD '23

Sponsor:

KDD '23: The 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 6 - 10, 2023

CA, Long Beach, USA

Acceptance Rates

Overall Acceptance Rate 1,133 of 8,635 submissions, 13%

Upcoming Conference

KDD '25

Sponsor:
sigkdd
sigkdd

The 31st ACM SIGKDD Conference on Knowledge Discovery and Data Mining

August 3 - 7, 2025

Toronto , ON , Canada

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

4
Total Citations
View Citations
678
Total Downloads

Downloads (Last 12 months)326
Downloads (Last 6 weeks)49

Reflects downloads up to 28 Feb 2025

Other Metrics

View Author Metrics

Citations

Cited By

Zhang ZWang XChen HLi HZhu W(2024)Disentangled Dynamic Graph Attention Network for Out-of-Distribution Sequential RecommendationACM Transactions on Information Systems10.1145/370198843:1(1-42)Online publication date: 29-Oct-2024
https://dl.acm.org/doi/10.1145/3701988
Lv ZHe SZhan TZhang SZhang WChen JZhao ZWu FCai JKankanhalli MPrabhakaran BBoll SSubramanian RZheng LSingh VCesar PXie LXu D(2024)Semantic Codebook Learning for Dynamic Recommendation ModelsProceedings of the 32nd ACM International Conference on Multimedia10.1145/3664647.3680574(9611-9620)Online publication date: 28-Oct-2024
https://dl.acm.org/doi/10.1145/3664647.3680574
Chen ZXiao TWang DZhang M(2024)Pareto Graph Self-Supervised LearningICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)10.1109/ICASSP48485.2024.10447557(6630-6634)Online publication date: 14-Apr-2024
https://doi.org/10.1109/ICASSP48485.2024.10447557
Tang MZou LCui SLiang SJin Z(2024)Unbiased Recommendation Through Invariant Representation LearningMachine Learning and Knowledge Discovery in Databases. Applied Data Science Track10.1007/978-3-031-70381-2_18(280-296)Online publication date: 22-Aug-2024
https://doi.org/10.1007/978-3-031-70381-2_18

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Figures

Tables

Media

View Table of Conten