Fairness-aware PAC learning from corrupted data

article Fairness-aware PAC learning from corrupted data published yes Nikola H Konstantinov author 4B9D76E4-F248-11E8-B48F-1D18A9856A870009-0009-5204-7621 Christoph Lampert author 40C20FD2-F248-11E8-B48F-1D18A9856A870000-0002-4561-241X ChLa department Addressing fairness concerns about machine learning models is a crucial step towards their long-term adoption in real-world automated systems. While many approaches have been developed for training fair models from data, little is known about the robustness of these methods to data corruption. In this work we consider fairness-aware learning under worst-case data manipulations. We show that an adversary can in some situations force any learner to return an overly biased classifier, regardless of the sample size and with or without degrading accuracy, and that the strength of the excess bias increases for learning problems with underrepresented protected groups in the data. We also prove that our hardness results are tight up to constant factors. To this end, we study two natural learning algorithms that optimize for both accuracy and fairness and show that these algorithms enjoy guarantees that are order-optimal in terms of the corruption ratio and the protected groups frequencies in the large data limit. https://research-explorer.ista.ac.at/download/10802/11570/2022_JournalMachineLearningResearch_Konstantinov.pdf application/pdfno ML Research Press2022 eng Fairnessrobustnessdata poisoningtrustworthy machine learningPAC learning Journal of Machine Learning Research 1532-4435 1533-7928 2102.06004 231-60 https://research-explorer.ista.ac.at/record/13241 https://research-explorer.ista.ac.at/record/10799 N. H. Konstantinov and C. Lampert, “Fairness-aware PAC learning from corrupted data,” Journal of Machine Learning Research, vol. 23. ML Research Press, pp. 1–60, 2022. Konstantinov NH, Lampert C. Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. 2022;23:1-60. Konstantinov NH, Lampert C. 2022. Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. 23, 1–60. N.H. Konstantinov, C. Lampert, Journal of Machine Learning Research 23 (2022) 1–60. Konstantinov, Nikola H., and Christoph Lampert. “Fairness-Aware PAC Learning from Corrupted Data.” Journal of Machine Learning Research, vol. 23, ML Research Press, 2022, pp. 1–60. Konstantinov, Nikola H, and Christoph Lampert. “Fairness-Aware PAC Learning from Corrupted Data.” Journal of Machine Learning Research. ML Research Press, 2022. Konstantinov, N. H., & Lampert, C. (2022). Fairness-aware PAC learning from corrupted data. Journal of Machine Learning Research. ML Research Press. 108022022-02-28T14:05:42Z2026-04-07T14:19:48Z