Abstract:
Fair regression methods typically rely on squared error loss, making them fragile under heavy tailed noise. We propose a general framework for robust regression under demographic parity (DP) that applies to a wide class of M-estimators, including Cauchy, Huber, least absolute deviation, quantile, and Tukey losses. We propose an optimal fair transformation that guarantees DP while achieving the minimum population risk among all rank preserving fair predictors. We also establish convergence rates for the resulting estimators. To balance fairness and predictive accuracy, we develop an interpolation scheme whose risk decreases while unfairness grows linearly with the interpolation parameter. The proposed framework can be further extended to conditional DP to account for legitimate covariates. Extensive simulation studies and real data applications show clear improvements over existing fair regression approaches in both robustness and predictive performance.
报告人简介:
Dr. Wen Su received her PhD in Statistics from the University of Hong Kong. She earned her Bachelor of Science in Industrial Engineering from the University of Toronto and Master of Science in Biostatistics from Columbia University. Her research interests include statistical machine learning, survival analysis, longitudinal data analysis, causal inference, and nonparametric and semiparametric inference. In survival analysis, she develops deep learning–based methods for complex time-to-event data, especially in medical applications. She also designs methods for optimal individualized treatment regimes to improve precision and effectiveness of care. She received the NSFC Young Scientist Grant (2024) and the RGC Early Career Scheme Grant (2025). She has published papers in prestigious journals including Journal of the Royal Statistical Society Series B, Biometrika, Biometrics, Bernoulli, Statistica Sinica, Electronic Journal of Statistics, Statistics in Medicine, etc. She serves as a reviewer for journals such as the Journal of the American Statistical Association, Journal of Machine Learning Research, Annals of Applied Statistics, Scandinavian Journal of Statistics, etc.
