Events

Math & Democracy: Understanding Last Layer Retraining Methods for Fair Classification: Theory and Algorithms

Speaker: Lalitha Sankar, Arizona State University

Location: 60 Fifth Avenue, Room 650

Date: Monday, February 24, 2025

Last-layer retraining (LLR) methods have emerged as an efficient framework for ensuring fairness and robustness in deep models. In this talk, we present an overview of existing methods and provide theoretical guarantees for several prominent methods. Under the threat of label noise, either in the class or domain annotations, we show that these naive methods fail. To address these issues, we present a new robust LLR method in the framework of two-stage corrections and demonstrate that it achieves state-of-the-art performance under domain label noise with minimal data overhead. Finally, we demonstrate that class label noise causes catastrophic failures even with robust two-stage methods, and propose a drop-in label correction which outperforms existing methods with very low computational and data cost.