Accelerated-gradient-based generalized Levenberg--Marquardt method with oracle complexity bound and local quadratic convergence

Abstract

Many machine learning tasks are formulated as a nonconvex optimization problem of minimizing the sum of a convex function and a composite function. The generalized Levenberg-Marquardt (LM) method, also known as the prox-linear method, has been developed for such problems. The method iteratively solves strongly convex subproblems with a damping term.This study proposes a new generalized LM method for solving the problem with a smooth composite function. The method enjoys three theoretical guarantees: iteration complexity bound, oracle complexity bound, and local convergence under a Hölderian growth condition. The local convergence results include local quadratic convergence under the quadratic growth condition; this is the first to extend a classical result for least-squares problems to a general smooth composite function. In addition, this is the first LM method with both an oracle complexity bound and local quadratic convergence under standard assumptions. These results are achieved by carefully controlling the damping parameter and solving the subproblems by the accelerated proximal gradient method equipped with a particular termination condition. Experimental results show that the proposed method performs well in practice for several instances, including classification with a neural network and collaborative filtering with nonnegative matrix factorization.