Tutorial on Fairness of Machine Learning in Recommender Systems

The tutorial is delivered at SIGIR 2021.

Abstract

Recently, there has been growing attention on fairness considerations in machine learning. As one of the most pervasive applications of machine learning, recommender systems are gaining increasing and critical impacts on human and society since a growing number of users use them for information seeking and decision making. Therefore, it is crucial to address the potential unfairness problems in recommendation, which may hurt users’ or providers’ satisfaction in recommender systems as well as the interests of the platforms. The tutorial focuses on the foundations and algorithms for fairness in recommendation. It also presents a brief introduction about fairness in basic machine learning tasks such as classification and ranking. The tutorial will introduce the taxonomies of current fairness definitions and evaluation metrics for fairness concerns. We will introduce previous works about fairness in recommendation and also put forward future fairness research directions. The tutorial aims at introducing and communicating fairness in recommendation methods to the community, as well as gathering researchers and practitioners interested in this research direction for discussions, idea communications, and research promotions.

Outline

  • Introduction
    • Social Impact of Recommender System and Fairness
    • Motivation of Fairness in Recommender Systems
    • Relationship with AI Ethics
    • Beyond Ethics: a Utilitarian Perspective
  • Fairness in Machine Learning
    • Fairness in Classification
    • Fairness in Ranking
  • Fairness in Recommendation
    • Introduction
    • Taxonomy
    • Dataset and Evaluation
    • Challenge and Opportunity

Material

PDF Slides

Presenters

Yunqi Li is a Ph.D. student in the Department of Computer Science at Rutgers University advised by Prof. Yongfeng Zhang. Her research interests lie in the intersection of Machine Learning and Information Retrieval. Her recent researches focus on AI Ethics including bringing fairness and interpretability to machine learning algorithms, as well as causal inference in machine learning. Her works have appeared in premier IR and AI/ML conferences such as WWW, SIGIR, WSDM, AAAI, etc.

Yingqiang Ge is a PhD student at the Computer Science Department of Rutgers University supervised by Prof. Yongfeng Zhang. His research interests broadly lie in IR and machine learning, including economic recommendation, explainable recommendation and fairness in recommendation, etc. His recent work on fairness includes fairness in explainable recommendation, long-term fairness in recommendation, user-oriented fairness, and fairness-aware IR evaluation. He has served as PC member/reviewer in top computer science conferences or journals such as KDD, SIGIR, IJCAI, AAAI, RecSys and ACM TOIS.

Yongfeng Zhang is an Assistant Professor in the Department of Computer Science at Rutgers University (The State University of New Jersey). His research interest is in Information Retrieval, Economics of Data Science, Explainable AI, Fairness in AI, and AI Ethics. In the previous he was a postdoc advised by Prof. W. Bruce Croft in the Center for Intelligent Information Retrieval (CIIR) at UMass Amherst, and did his PhD and BE in Computer Science at Tsinghua University, with a BS in Economics at Peking University. He is a Siebel Scholar of the class 2015, and a Baidu Scholar of the class 2014. Together with coauthors, he has been consistently working on explainable search and recommendation models, fairness-aware recommendation, echo chambers in IR systems, as well as causal/counterfactual models for information retrieval. His recent research on fairness in recommendation include long-term fairness, useroriented fairness, group fairness, explainable fairness, Pareto fairness and fairness/diversity in echo chambers. He has served as PC members or senior PC members in various Web&IR related conferences such as SIGIR, WWW, CIKM, WSDM, ICTIR and CHIIR, and he is serving as the associate editor for ACM Transactions on Information Systems (TOIS). He has presented the WWW’19/SIGIR’19/ICTIR’19 Tutorial on Explainable Recommendation and Search, and the RecSys’20/WSDM’21 Tutorial on Conversational Recommendation.