Robustness and Generalization

We derive generalization bounds for learning algorithms based on their robustness: the property that if a testing sample is "similar" to a training sample, then the testing error is close to the training error. This provides a novel approach, different from the complexity or stability arguments, to study generalization of learning algorithms. We further show that a weak notion of robustness is both sufficient and necessary for generalizability, which implies that robustness is a fundamental property for learning algorithms to work.

• Type: Journal
• Year: 2010
• Where: CORR
• Authors: Huan Xu, Shie Mannor
• Keywords: Algorithms, CORR 2010, Education, Generalization Bounds, Robustness