Generalization Error Bounds for Optimization Algorithms via Stability

TL;DR

This paper derives bounds on the generalization error of optimization algorithms like GD, SGD, and SVRG using stability, showing how their convergence impacts test performance in convex and non-convex settings.

Contribution

It provides the first theoretical analysis linking optimization convergence rates with generalization error bounds via stability for both convex and non-convex problems.

Findings

Generalization error decreases with training for all algorithms studied.

SVRG exhibits better generalization ability than GD and SGD.

Experimental results confirm the theoretical bounds and insights.

Abstract

Many machine learning tasks can be formulated as Regularized Empirical Risk Minimization (R-ERM), and solved by optimization algorithms such as gradient descent (GD), stochastic gradient descent (SGD), and stochastic variance reduction (SVRG). Conventional analysis on these optimization algorithms focuses on their convergence rates during the training process, however, people in the machine learning community may care more about the generalization performance of the learned model on unseen test data. In this paper, we investigate on this issue, by using stability as a tool. In particular, we decompose the generalization error for R-ERM, and derive its upper bound for both convex and non-convex cases. In convex cases, we prove that the generalization error can be bounded by the convergence rate of the optimization algorithm and the stability of the R-ERM process, both in expectation (in the order of $\mathcal{O}((1/n)+\mathbb{E}\rho(T))$, where $\rho(T)$ is the convergence error and $T$ is the number of iterations) and in high probability (in the order of $\mathcal{O}\left(\frac{\log{1/\delta}}{\sqrt{n}}+\rho(T)\right)$ with probability $1-\delta$). For non-convex cases, we can also obtain a similar expected generalization error bound. Our theorems indicate that 1) along with the training process, the generalization error will decrease for all the optimization algorithms under our investigation; 2) Comparatively speaking, SVRG has better generalization ability than GD and SGD. We have conducted experiments on both convex and non-convex problems, and the experimental results verify our theoretical findings.