Huimin Zeng's Final Defense
PhD candidate Huimin Zeng will present his dissertation defense, “Trustworthy Representation Learning in Federated and Multi-modal Foundation Models.” Zeng's dissertation committee includes Professor J. Stephen Downie (Chair), Professor Jingrui He, Associate Professor Halil Kilicoglu, Assistant Professor Madelyn Rose Sanfilippo, and Associate Professor Dimitrios Katselis.
Abstract
The rapid advancement of machine learning (ML) has enabled high-impact applications across domains. However, existing ML systems still exhibit undesired behaviors (e.g., demographic discrimination, privacy risks, and model vulnerability under distribution shift), raising concerns about their trustworthiness. This thesis explores the trustworthiness of ML systems through the lens of trustworthy representation learning, with the ultimate goal of developing principled and practical solutions to improve trustworthiness of ML systems, such as fairness, privacy, robustness in federated and multimodal foundation models.
Questions? Contact Huimin Zeng.