Xinying Zou - 邹馨莹
Xinying Zou - 邹馨莹
Dec. 2022 - Feb. 2026, PhD student at Centre Inria d'Université Côte d'Azur
under the supervision of Samir M. Perlaza, Eitan Altman, and Iñaki Esnaola.
Contact: xinyingzou321@gmail.com
Embedded Files
Research interests:
Research interests:
Information Theory, Statistical Learning.
Research Overview:
During my Ph.D., I am interested in the generalization theory of machine learning algorithms. My Ph.D. thesis develops an information-theoretic framework to characterize and improve the generalization capabilities of supervised learning algorithms. Through distributionally robust optimization with Kullback-Leibler divergence constraints, closed-form expressions for generalization error and model sensitivity are provided in terms of information-theoretic measures. Building on a training-dependent minimax framework where distribution shifts are modeled under a KL-divergence constraint, an algorithm robustification method is proposed to improve the performance of a given learning algorithm with respect to unseen data, which addresses the gap between empirical training performance and reliability under distribution shifts.
Page updated
Google Sites
Report abuse