Timezone: »

Fairness-Aware PAC Learning from Corrupted Data
Nikola Konstantinov · Christoph Lampert

Wed Nov 30 02:00 PM -- 04:00 PM (PST) @ Hall J #1010

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.

Author Information

Nikola Konstantinov (ETH Zurich)
Christoph Lampert (IST Austria)
Christoph Lampert

Christoph Lampert received the PhD degree in mathematics from the University of Bonn in 2003. In 2010 he joined the Institute of Science and Technology Austria (ISTA) first as an Assistant Professor and since 2015 as a Professor. There, he leads the research group for Machine Learning and Computer Vision, and since 2019 he is also the head of ISTA's ELLIS unit.

More from the Same Authors