"And the Winner Is...": Dynamic Lotteries for Multi-group Fairness-Aware Recommendation

As recommender systems are being designed and deployed for an increasing number of socially-consequential applications, it has become important to consider what properties of fairness these systems exhibit. There has been considerable research on recommendation fairness. However, we argue that the previous literature has been based on simple, uniform and often uni-dimensional notions of fairness assumptions that do not recognize the real-world complexities of fairness-aware applications. In this paper, we explicitly represent the design decisions that enter into the trade-off between accuracy and fairness across multiply-defined and intersecting protected groups, supporting multiple fairness metrics. The framework also allows the recommender to adjust its performance based on the historical view of recommendations that have been delivered over a time horizon, dynamically rebalancing between fairness concerns. Within this framework, we formulate lottery-based mechanisms for choosing between fairness concerns, and demonstrate their performance in two recommendation domains.

[1]  R. Burke,et al.  Algorithmic Fairness, Institutional Logics, and Social Choice , 2020 .

[2]  Nathan Srebro,et al.  Equality of Opportunity in Supervised Learning , 2016, NIPS.

[3]  A. Sen,et al.  Collective Choice and Social Welfare , 2017 .

[4]  Haris Aziz Developments in Multi-Agent Fair Allocation , 2020, AAAI.

[5]  Alexandra Chouldechova,et al.  Fair prediction with disparate impact: A study of bias in recidivism prediction instruments , 2016, Big Data.

[6]  P. Rousseeuw Silhouettes: a graphical aid to the interpretation and validation of cluster analysis , 1987 .

[7]  Weiwen Liu,et al.  Personalizing Fairness-aware Re-ranking , 2018, ArXiv.

[8]  Bert Huang,et al.  Beyond Parity: Fairness Objectives for Collaborative Filtering , 2017, NIPS.

[9]  Krishna P. Gummadi,et al.  Two-Sided Fairness for Repeated Matchings in Two-Sided Markets: A Case Study of a Ride-Hailing Platform , 2019, KDD.

[10]  Robin Burke,et al.  Opportunistic Multi-aspect Fairness through Personalized Re-ranking , 2020, UMAP.

[11]  Toniann Pitassi,et al.  Fairness through awareness , 2011, ITCS '12.

[12]  Nicholas Rescher,et al.  Fairness: Theory & Practice of Distributive Justice , 2002 .

[13]  William Thomson,et al.  Introduction to the Theory of Fair Allocation , 2016, Handbook of Computational Social Choice.

[14]  Lu Zhang,et al.  Anti-discrimination learning: a causal modeling-based framework , 2017, International Journal of Data Science and Analytics.

[15]  Lior Rokach,et al.  Data Mining And Knowledge Discovery Handbook , 2005 .

[16]  Seth Neel,et al.  Preventing Fairness Gerrymandering: Auditing and Learning for Subgroup Fairness , 2017, ICML.

[17]  Alvin E. Roth,et al.  Pairwise Kidney Exchange , 2004, J. Econ. Theory.

[18]  Dan W. Brockt,et al.  The Theory of Justice , 2017 .

[19]  Nasim Sonboli,et al.  Balanced Neighborhoods for Multi-sided Fairness in Recommendation , 2018, FAT.

[20]  Robin D. Burke,et al.  Multisided Fairness for Recommendation , 2017, ArXiv.

[21]  Bamshad Mobasher,et al.  A Clustering Approach for Personalizing Diversity in Collaborative Recommender Systems , 2017, UMAP.

[22]  Nicholas Mattei,et al.  Egalitarianism of Random Assignment Mechanisms , 2015, ArXiv.

[23]  Jun Sakuma,et al.  Enhancement of the Neutrality in Recommendation , 2012, Decisions@RecSys.

[24]  Fernando Diaz,et al.  Towards a Fair Marketplace: Counterfactual Evaluation of the trade-off between Relevance, Fairness & Satisfaction in Recommendation Systems , 2018, CIKM.

[25]  Vincent Conitzer,et al.  Handbook of Computational Social Choice , 2016 .

[26]  Nicholas Mattei,et al.  Closing the Loop: Bringing Humans into Empirical Computational Social Choice and Preference Reasoning , 2020, IJCAI.

[27]  Shotaro Akaho,et al.  Model-Based Approaches for Independence-Enhanced Recommendation , 2016, 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW).

[28]  H. Peyton Young,et al.  Equity - in theory and practice , 1994 .

[29]  Hervé Moulin,et al.  A New Solution to the Random Assignment Problem , 2001, J. Econ. Theory.

[30]  Guy N. Rothblum,et al.  Multicalibration: Calibration for the (Computationally-Identifiable) Masses , 2018, ICML.

[31]  Michael D. Ekstrand,et al.  Exploring author gender in book rating and recommendation , 2018, User Modeling and User-Adapted Interaction.

[32]  David Manlove,et al.  Algorithmics of Matching Under Preferences , 2013, Bull. EATCS.

[33]  Olfa Nasraoui,et al.  Debiasing the Human-Recommender System Feedback Loop in Collaborative Filtering , 2019, WWW.

[34]  Ariel D. Procaccia,et al.  WeBuildAI , 2019, Proc. ACM Hum. Comput. Interact..

[35]  Toniann Pitassi,et al.  Learning Fair Representations , 2013, ICML.

[36]  Dietmar Jannach,et al.  Multistakeholder recommendation: Survey and research directions , 2020, User Modeling and User-Adapted Interaction.

[37]  Krishna P. Gummadi,et al.  Equality of Voice: Towards Fair Representation in Crowdsourced Top-K Recommendations , 2018, FAT.

[38]  Jun Sakuma,et al.  Recommendation Independence , 2018, FAT.

[39]  Bamshad Mobasher,et al.  Managing Popularity Bias in Recommender Systems with Personalized Re-ranking , 2019, FLAIRS.

[40]  Franco Turini,et al.  Discrimination-aware data mining , 2008, KDD.

[41]  Ed H. Chi,et al.  Fairness in Recommendation Ranking through Pairwise Comparisons , 2019, KDD.

[42]  Domonkos Tikk,et al.  Investigation of Various Matrix Factorization Methods for Large Recommender Systems , 2008, 2008 IEEE International Conference on Data Mining Workshops.

[43]  Gediminas Adomavicius,et al.  Working Paper: Improving Recommendation Diversity Using Ranking-based Techniques Improving Recommendation Diversity Using Ranking-based Techniques , 2022 .

[44]  Jun Sakuma,et al.  Fairness-Aware Classifier with Prejudice Remover Regularizer , 2012, ECML/PKDD.

[45]  Hervé Moulin,et al.  Fair division and collective welfare , 2003 .

[46]  Paul R. Milgrom,et al.  Designing Random Allocation Mechanisms: Theory and Applications , 2013 .

[47]  Krishna P. Gummadi,et al.  FairRec: Two-Sided Fairness for Personalized Recommendations in Two-Sided Platforms , 2020, WWW.

[48]  James Caverlee,et al.  Fairness-Aware Tensor-Based Recommendation , 2018, CIKM.

[49]  Toby Walsh,et al.  The Conference Paper Assignment Problem: Using Order Weighted Averages to Assign Indivisible Goods , 2017, AAAI.

[50]  Toby Walsh,et al.  Equilibria Under the Probabilistic Serial Rule , 2015, IJCAI.

[51]  Maria Soledad Pera,et al.  All The Cool Kids, How Do They Fit In?: Popularity and Demographic Biases in Recommender Evaluation and Effectiveness , 2018, FAT.

[52]  W. Thomson Fair Allocation Rules , 2011 .

[53]  Timnit Gebru,et al.  Gender Shades: Intersectional Accuracy Disparities in Commercial Gender Classification , 2018, FAT.

[54]  Jun Guo,et al.  Personalized fairness-aware re-ranking for microlending , 2019, RecSys.

[55]  William S. Zwicker,et al.  Introduction to the Theory of Voting , 2016, Handbook of Computational Social Choice.

[56]  Amit Sharma,et al.  FATREC Workshop on Responsible Recommendation , 2017, RecSys.

[57]  Lior Rokach,et al.  Clustering Methods , 2005, The Data Mining and Knowledge Discovery Handbook.

[58]  Toon Calders,et al.  Discrimination Aware Decision Tree Learning , 2010, 2010 IEEE International Conference on Data Mining.