Towards Optimal Problem Dependent Generalization Error Bounds in Statistical Learning Theory

Abstract

We study problem-dependent rates, i.e., generalization errors that scale near-optimally with the variance, the effective loss, or the gradient norms evaluated at the "best hypothesis." We introduce a principled framework dubbed "uniform localized convergence," and characterize sharp problem-dependent rates for central statistical learning problems. From a methodological viewpoint, our framework resolves several fundamental limitations of existing uniform convergence and localization analysis approaches. It also provides improvements and some level of unification in the study of localized complexities, one-sided uniform inequalities, and sample-based iterative algorithms. In the so-called "slow rate" regime, we provides the first (moment-penalized) estimator that achieves the optimal variance-dependent rate for general "rich" classes; we also establish improved loss-dependent rate for standard empirical risk minimization. In the "fast rate" regime, we establish finite-sample problem-dependent bounds that are comparable to precise asymptotics. In addition, we show that iterative algorithms like gradient descent and first-order Expectation-Maximization can achieve optimal generalization error in several representative problems across the areas of non-convex learning, stochastic optimization, and learning with missing data.

References in corpus (5)

• Empirical Bernstein Bounds and Sample Variance Penalization
• Counterfactual Risk Minimization: Learning from Logged Bandit Feedback
• Logistic Regression: Tight Bounds for Stochastic and Online Optimization
• High Dimensional Expectation-Maximization Algorithm: Statistical Optimization and Asymptotic Normality
• SGD Converges to Global Minimum in Deep Learning via Star-convex Path