Distributional Robustness Bounds Generalization Errors

arXiv:2212.09962v3 Announce Type: replace
Abstract: Bayesian methods, distributionally robust optimization methods, and regularization methods are three pillars of trustworthy machine learning combating distributional uncertainty, e.g., the uncertainty of an empirical distribution compared to the true underlying distribution. This paper investigates the connections among the three frameworks and, in particular, explores why these frameworks tend to have smaller generalization errors. Specifically, first, we suggest a quantitative definition for "distributional robustness", propose the concept of "robustness measure", and formalize several philosophical concepts …

abstract arxiv bayesian cs.lg distribution errors frameworks machine machine learning optimization paper regularization robust robustness stat.ml true trustworthy type uncertainty