A New Convergence Analysis of Two Stochastic Frank-Wolfe Algorithms
Abstract
We study the convergence properties of the original and away-step Frank-Wolfe algorithms for linearly constrained stochastic optimization assuming the availability of unbiased objective function gradient estimates. The objective function is not restricted to a finite summation form, like in previous analyses tailored to machine-learning applications. To enable the use of concentration inequalities we assume either a uniform bound on the variance of gradient estimates or uniformly sub-Gaussian tails on gradient estimates. With one of these regularity assumptions along with sufficient sampling, we can ensure sufficiently accurate gradient estimates. We then use a Lyapunov argument to obtain the desired complexity bounds, relying on existing geometrical results for polytopes.
Cite
@article{arxiv.2504.04213,
title = {A New Convergence Analysis of Two Stochastic Frank-Wolfe Algorithms},
author = {Natthawut Boonsiriphatthanajaroen and Shane G. Henderson},
journal= {arXiv preprint arXiv:2504.04213},
year = {2025}
}