Complexity guarantees for an implicit smoothing-enabled method for stochastic MPECs

TL;DR

This paper introduces a zeroth-order implicit smoothing algorithm for stochastic MPECs, providing non-asymptotic complexity guarantees for both convex and nonconvex cases, with extensive theoretical and numerical validation.

Contribution

It develops a novel smoothing-based zeroth-order method with explicit complexity bounds for stochastic MPECs, addressing a gap in efficient algorithms with guarantees.

Findings

Complexity bounds are established for single-stage and two-stage SMPECs.

The proposed schemes achieve non-asymptotic convergence guarantees.

Numerical results validate the theoretical complexity bounds.

Abstract

Stochastic MPECs have found increasing relevance for modeling a broad range of settings in engineering and statistics. Yet, there seem to be no efficient first/zeroth-order schemes equipped with non-asymptotic rate guarantees for resolving even deterministic variants of such problems. We consider SMPECs where the parametrized lower-level equilibrium problem is given by a deterministic/stochastic VI problem whose mapping is strongly monotone. We develop a zeroth-order implicit algorithmic framework by leveraging a locally randomized spherical smoothing scheme. We present schemes for single-stage and two-stage stochastic MPECs when the upper-level problem is either convex or nonconvex. (I). Single-stage SMPECs: In convex regimes, our proposed inexact schemes are characterized by a complexity in upper-level projections, upper-level samples, and lower-level projections of $\mathcal{O}(\tfrac{1}{\epsilon^2})$, $\mathcal{O}(\tfrac{1}{\epsilon^2})$, and $\mathcal{O}(\tfrac{1}{\epsilon^2}\ln(\tfrac{1}{\epsilon}))$ , respectively. Analogous bounds for the nonconvex regime are $\mathcal{O}(\tfrac{1}{\epsilon})$, $\mathcal{O}(\tfrac{1}{\epsilon^2})$, and $\mathcal{O}(\tfrac{1}{\epsilon^3})$, respectively . (II). Two-stage SMPECs: In convex regimes, our proposed inexact schemes have a complexity in upper-level projections, upper-level samples, and lower-level projections of $\mathcal{O}(\tfrac{1}{\epsilon^2}),\mathcal{O}(\tfrac{1}{\epsilon^2})$, and $\mathcal{O}(\tfrac{1}{\epsilon^2}\ln(\tfrac{1}{\epsilon}))$ while the corresponding bounds in the nonconvex regime are $\mathcal{O}(\tfrac{1}{\epsilon})$, $\mathcal{O}(\tfrac{1}{\epsilon^2})$, and $\mathcal{O}(\tfrac{1}{\epsilon^2}\ln(\tfrac{1}{\epsilon}))$ , respectively . In addition, we derive statements for exact as well as accelerated counterparts. We also provide a comprehensive set of numerical results for validating the theoretical findings.