Stochastic Approximation of Smooth and Strongly Convex Functions: Beyond the $O(1/T)$ Convergence Rate
This work provides faster convergence guarantees for stochastic convex optimization, which is incremental but important for machine learning and optimization practitioners.
The paper tackles the problem of improving the convergence rate of stochastic approximation for smooth and strongly convex functions, achieving risk bounds of O(1/[λT^α] + κF_*/T) and O(1/2^{T/κ} + F_*), with exponential reduction to O(F_*) when the minimal risk is small.
Stochastic approximation (SA) is a classical approach for stochastic convex optimization. Previous studies have demonstrated that the convergence rate of SA can be improved by introducing either smoothness or strong convexity condition. In this paper, we make use of smoothness and strong convexity simultaneously to boost the convergence rate. Let $λ$ be the modulus of strong convexity, $κ$ be the condition number, $F_*$ be the minimal risk, and $α>1$ be some small constant. First, we demonstrate that, in expectation, an $O(1/[λT^α] + κF_*/T)$ risk bound is attainable when $T = Ω(κ^α)$. Thus, when $F_*$ is small, the convergence rate could be faster than $O(1/[λT])$ and approaches $O(1/[λT^α])$ in the ideal case. Second, to further benefit from small risk, we show that, in expectation, an $O(1/2^{T/κ}+F_*)$ risk bound is achievable. Thus, the excess risk reduces exponentially until reaching $O(F_*)$, and if $F_*=0$, we obtain a global linear convergence. Finally, we emphasize that our proof is constructive and each risk bound is equipped with an efficient stochastic algorithm attaining that bound.