Performative Debias with Fair-exposure Optimization Driven by Strategic Agents in Recommender Systems
Not on the map. Why?
Abstract (arXiv)
Data bias, e.g., popularity impairs the dynamics of two-sided markets within recommender systems. This overshadows the less visible but potentially intriguing long-tail items that could capture user interest. Despite the abundance of research surrounding this issue, it still poses challenges and remains a hot topic in academic circles. Along this line, in this paper, we developed a re-ranking approach in dynamic settings with fair-exposure optimization driven by strategic agents. Designed for the producer side, the execution of agents assumes content creators can modify item features based on strategic incentives to maximize their exposure. This iterative process entails an end-to-end optimization, employing differentiable ranking operators that simultaneously target accuracy and fairness. Joint objectives ensure the performance of recommendations while enhancing the visibility of tail items. We also leveraged the performativity nature of predictions to illustrate how strategic learning influences content creators to shift towards fairness efficiently, thereby incentivizing features of tail items. Through comprehensive experiments on both public and industrial datasets, we have substantiated the effectiveness and dominance of the proposed method especially on unveiling the potential of tail items.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 30 of its 55 references that are in the map. First posted on arXiv 25 Jun 2024.
Builds on
30 of 55 references.
References · 30
- 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,739
- 2020Performative Predictioncited by 327
- 2015Strategic Classificationcited by 257
- 2016Recommendations as Treatments: Debiasing Learning and Evaluationcited by 248
- 2020Bias and Debias in Recommender System: A Survey and Future Directionscited by 240
- 2017How Algorithmic Confounding in Recommendation Systems Increases Homogeneity and Decreases Utilitycited by 120
- 2021Causal Intervention for Leveraging Popularity Bias in Recommendationcited by 116
- 2019Stochastic Optimization of Sorting Networks via Continuous Relaxationscited by 111
- 2020Model-Agnostic Counterfactual Reasoning for Eliminating Popularity Bias in Recommender Systemcited by 94
- 2020Feedback Loop and Bias Amplification in Recommender Systemscited by 90
- 2020Performative Prediction in a Stateful Worldcited by 90
- 2019Managing Popularity Bias in Recommender Systems with Personalized Re-rankingcited by 82
18 more
- 2020FairRec: Two-Sided Fairness for Personalized Recommendations in Two-Sided Platformscited by 78
- 2020Controlling Fairness and Bias in Dynamic Learning-to-Rankcited by 68
- 2019Degenerate Feedback Loops in Recommender Systemscited by 67
- 2022Data Feedback Loops: Model-driven Amplification of Dataset Biasescited by 60
- 2023Machine Culturecited by 55
- 2023Exploring Large Language Model for Graph Data Understanding in Online Job Recommendationscited by 45
- 2021NeuralNDCG: Direct Optimisation of a Ranking Metric via Differentiable Relaxation of Sortingcited by 41
- 2021User-centered Evaluation of Popularity Bias in Recommender Systemscited by 38
- 2022Anticipating Performativity by Predicting from Predictionscited by 37
- 2022Modeling Content Creator Incentives on Algorithm-Curated Platformscited by 35
- 2018Reducing Disparate Exposure in Ranking: A Learning To Rank Approachcited by 33
- 2022CPFair: Personalized Consumer and Producer Fairness Re-ranking for Recommender Systemscited by 29
- 2020From Predictions to Decisions: Using Lookahead Regularizationcited by 19
- 2021End-to-end Learning for Fair Ranking Systemscited by 18
- 2023Improving Recommendation Fairness via Data Augmentationcited by 18
- 2022Optimizing generalized Gini indices for fairness in rankingscited by 13
- 2023Performative Recommendation: Diversifying Content via Strategic Incentivescited by 12
- 2022Multi-Label Learning to Rank through Multi-Objective Optimizationcited by 10
Built on by
Citing papers per year: 2024: 1; 2025: 2.
3 papers.