webinar register page

Machine Learning Center Seminar: Yang Liu (UC Santa Cruz) - Agency Bias in Machine Learning

Abstract: A trained machine learning model (e.g., a classifier) will ultimately observe data generated according to agents' responses. For instance, the rising literature on strategic classification concerns the setting where agents are fully rational and can best respond to a classifier in their own interests. The above interaction will lead to a distribution shift between training and deployment and will challenge the existing performance and fairness guarantees of the trained model. In this talk, I'll discuss three types of agency bias that arise due to the above interactional effects between agents and machine learning models. I'll then go over possible mitigation efforts, including our very recent works on certifying the fairness guarantees on an unknown and possibly different deployment distribution.

References:
[1] Unfairness Despite Awareness: Group-Fair Classification with Strategic Agents. Andrew Estornell, Sanmay Das, Yang Liu and Yevgeniy Vorobeychik. Preprint, 2022.
[2] Actionable Recourse in Linear Classification. Berk Ustun, Alexander Spangher and Yang Liu
ACM Conference on Fairness, Accountability, and Transparency (ACM FAccT), 2019.
[3] Unintended Selection: Persistent Qualification Rate Disparities and Interventions. Reilly Raab and Yang Liu. Neural Information Processing Systems (NeurIPS), 2021.
[4] Fairness Transferability Subject to Bounded Distribution Shift. Yatong Chen, Reilly Raab, Jialu Wang and Yang Liu. Neural Information Processing Systems (NeurIPS), 2022.

This lecture satisfies requirements for CSCI 591: Research Colloquium.

Nov 17, 2022 10:00 AM in Pacific Time (US and Canada)