TY - GEN
T1 - Learning Fair Ranking Policies via Differentiable Optimization of Ordered Weighted Averages
AU - Dinh, My H.
AU - Kotary, James
AU - Fioretto, Ferdinando
N1 - Publisher Copyright:
© 2024 Owner/Author.
PY - 2024/6/3
Y1 - 2024/6/3
N2 - Learning to Rank (LTR) is one of the most widely used machine learning applications. It is a key component in platforms with profound societal impacts, including job search, healthcare information retrieval, and social media content feeds. Conventional LTR models have been shown to produce biases results, stimulating a discourse on how to address the disparities introduced by ranking systems that solely prioritize user relevance. However, while several models of fair learning to rank have been proposed, they suffer from deficiencies either in accuracy or efficiency, thus limiting their applicability to real-world ranking platforms. This paper shows how efficiently-solvable fair ranking models, based on the optimization of Ordered Weighted Average (OWA) functions, can be integrated into the training loop of an LTR model to achieve favorable balances between fairness, user utility, and runtime efficiency. In particular, this paper is the first to show how to backpropagate through constrained optimizations of OWA objectives, enabling their use in integrated prediction and decision models.
AB - Learning to Rank (LTR) is one of the most widely used machine learning applications. It is a key component in platforms with profound societal impacts, including job search, healthcare information retrieval, and social media content feeds. Conventional LTR models have been shown to produce biases results, stimulating a discourse on how to address the disparities introduced by ranking systems that solely prioritize user relevance. However, while several models of fair learning to rank have been proposed, they suffer from deficiencies either in accuracy or efficiency, thus limiting their applicability to real-world ranking platforms. This paper shows how efficiently-solvable fair ranking models, based on the optimization of Ordered Weighted Average (OWA) functions, can be integrated into the training loop of an LTR model to achieve favorable balances between fairness, user utility, and runtime efficiency. In particular, this paper is the first to show how to backpropagate through constrained optimizations of OWA objectives, enabling their use in integrated prediction and decision models.
KW - Algorithmic Fairness
KW - Learning To Rank
KW - Predict-then-optimize
UR - http://www.scopus.com/inward/record.url?scp=85196626784&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85196626784&partnerID=8YFLogxK
U2 - 10.1145/3630106.3661932
DO - 10.1145/3630106.3661932
M3 - Conference contribution
AN - SCOPUS:85196626784
T3 - 2024 ACM Conference on Fairness, Accountability, and Transparency, FAccT 2024
SP - 2508
EP - 2517
BT - 2024 ACM Conference on Fairness, Accountability, and Transparency, FAccT 2024
PB - Association for Computing Machinery, Inc
T2 - 2024 ACM Conference on Fairness, Accountability, and Transparency, FAccT 2024
Y2 - 3 June 2024 through 6 June 2024
ER -