GLL-type Nonmonotone Descent Methods Revisited under Kurdyka-Łojasiewicz Property

TL;DR

The paper tackles the lack of full convergence theory for GLL-type nonmonotone methods in nonconvex, nonsmooth optimization by introducing a general iterative framework that couples a GLL descent rule with KL-based relative error and prox-regularity–driven gap conditions. It proves that any bounded pair of sequences following the framework converges, and that the convergence rate is linear or sublinear depending on the KL exponent $θ$. The authors show that this framework applies to broad KL composite problems and, importantly, provides full convergence guarantees for two established algorithms, NPG_major and PGenls/PGnls, without requiring gradient Lipschitz continuity. By linking prox-regularity and KL properties, the work broadens applicability to DC programs and other nonconvex, nonsmooth settings, offering a principled path for designing convergent GLL-type methods. Overall, the framework significantly strengthens the theoretical foundation for nonmonotone descent approaches in modern nonconvex optimization.

Abstract

The purpose of this paper is to extend the full convergence results of the classic GLL-type (Grippo-Lampariello-Lucidi) nonmonotone methods to nonconvex and nonsmooth optimization. We propose a novel iterative framework for the minimization of a proper and lower semicontinuous function $Φ$. The framework consists of the GLL-type nonmonotone decrease condition for a sequence, a relative error condition for its augmented sequence with respect to a Kurdyka-Łojasiewicz (KL) function $Θ$, and a relative gap condition for the partial maximum objective value sequence. The last condition is shown to be a product of the prox-regularity of $Φ$ on the set of cluster points, and to hold automatically under a mild condition on the objective value sequence. We prove that for any sequence and its bounded augmented sequence together falling within the framework, the sequence itself is convergent. Furthermore, when $Θ$ is a KL function of exponent $θ\in(0, 1)$, the convergence admits a linear rate if $θ\in(0, 1/2]$ and a sublinear rate if $θ\in(1/2, 1)$. As applications, we prove, for the first time, that the two existing algorithms, namely the nonmonotone proximal gradient (NPG) method with majorization and NPG with extrapolation both enjoy the full convergence of the iterate sequences for nonconvex and nonsmooth KL composite optimization problems.

Theorems & Definitions (37)

• Definition 1
• Definition 2
• Lemma 1
• Lemma 2
• Lemma 3
• proof
• Proposition 1
• proof
• Proposition 2
• proof