涩里番

Title: Generalization bounds via regret analysis鈥�

Abstract: Understanding the generalization ability of learning algorithms has been a key driving force behind statistical learning theory.

In this talk, we present a novel framework for deriving bounds on the generalization error of statistical learning algorithms from the perspective of online learning. Specifically, we construct an online learning game called the 鈥済eneralization game鈥�, where an online learner competes with a fixed statistical learning algorithm in predicting the sequence of generalization gaps on a training set of i.i.d. data points. We establish a connection between the online and statistical learning setting by showing that the existence of an online learning algorithm with bounded regret in this game implies a bound on the generalization error of the statistical learning algorithm. This technique allows us to recover several standard generalization bounds, including a range of PAC-Bayesian and information-theoretic guarantees.