Deterministic and Stochastic Frank-Wolfe Recursion on Probability Spaces

TL;DR

This paper develops deterministic and stochastic Frank-Wolfe algorithms for optimization over probability measures, providing convergence guarantees and practical implementations for applications like emergency response and experimental design.

Contribution

It introduces a novel deterministic Frank-Wolfe recursion for probability spaces and a stochastic variant that handles Monte Carlo observations, with proven convergence rates and a CLT for objective values.

Findings

sFW achieves $O(k^{-1})$ convergence for convex objectives

sFW achieves $O(k^{-1/2})$ convergence for non-convex objectives

Fixed-step sFW converges exponentially to $oldsymbol{ ext{ε}}$-optimality

Abstract

Motivated by applications in emergency response and experimental design, we consider smooth stochastic optimization problems over probability measures supported on compact subsets of the Euclidean space. With the influence function as the variational object, we construct a deterministic Frank-Wolfe (dFW) recursion for probability spaces, made especially possible by a lemma that identifies a ``closed-form'' solution to the infinite-dimensional Frank-Wolfe sub-problem. Each iterate in dFW is expressed as a convex combination of the incumbent iterate and a Dirac measure concentrating on the minimum of the influence function at the incumbent iterate. To address common application contexts that have access only to Monte Carlo observations of the objective and influence function, we construct a stochastic Frank-Wolfe (sFW) variation that generates a random sequence of probability measures constructed using minima of increasingly accurate estimates of the influence function. We demonstrate that sFW's optimality gap sequence exhibits $O(k^{-1})$ iteration complexity almost surely and in expectation for smooth convex objectives, and $O(k^{-1/2})$ (in Frank-Wolfe gap) for smooth non-convex objectives. Furthermore, we show that an easy-to-implement fixed-step, fixed-sample version of (sFW) exhibits exponential convergence to $\varepsilon$-optimality. We end with a central limit theorem on the observed objective values at the sequence of generated random measures. To further intuition, we include several illustrative examples with exact influence function calculations.