A General Recipe for Parameter-Free Nonconvex Optimization via Higher-Order Regularization

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Marumo, Naoki, Takeda, Akiko
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916064499073024
author Marumo, Naoki
Takeda, Akiko
author_facet Marumo, Naoki
Takeda, Akiko
contents We develop a systematic framework for constructing parameter-free algorithms for smooth nonconvex optimization. The framework is based on higher-order regularization: each step is computed from a regularized local model whose regularization exponent exceeds the order of the model error. This design makes the resulting method robust to misspecification of the regularization parameter and yields complexity bounds without backtracking or other acceptance tests. We apply the framework to gradient descent, Newton's method, the Gauss--Newton method, stochastic gradient descent, and PAGE. Without prior knowledge of problem-dependent parameters, the resulting algorithms achieve complexity bounds with optimal or best-known dependence on the target accuracy. When the problem-dependent parameters are known up to constant factors, suitable tuning also recovers the optimal or best-known dependence on those parameters.
format Preprint
id arxiv_https___arxiv_org_abs_2605_30891
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A General Recipe for Parameter-Free Nonconvex Optimization via Higher-Order Regularization
Marumo, Naoki
Takeda, Akiko
Optimization and Control
90C26, 90C30, 65K05, 49M15
We develop a systematic framework for constructing parameter-free algorithms for smooth nonconvex optimization. The framework is based on higher-order regularization: each step is computed from a regularized local model whose regularization exponent exceeds the order of the model error. This design makes the resulting method robust to misspecification of the regularization parameter and yields complexity bounds without backtracking or other acceptance tests. We apply the framework to gradient descent, Newton's method, the Gauss--Newton method, stochastic gradient descent, and PAGE. Without prior knowledge of problem-dependent parameters, the resulting algorithms achieve complexity bounds with optimal or best-known dependence on the target accuracy. When the problem-dependent parameters are known up to constant factors, suitable tuning also recovers the optimal or best-known dependence on those parameters.
title A General Recipe for Parameter-Free Nonconvex Optimization via Higher-Order Regularization
topic Optimization and Control
90C26, 90C30, 65K05, 49M15
url https://arxiv.org/abs/2605.30891