Fairness-Aware PAC Learning from Corrupted Data

Understand

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.