Loss-Transformation Invariance in the Damped Newton Method

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Shestakov, Alexander, Bohara, Sushil, Horváth, Samuel, Takáč, Martin, Hanzely, Slavomír
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916979360661504
author Shestakov, Alexander
Bohara, Sushil
Horváth, Samuel
Takáč, Martin
Hanzely, Slavomír
author_facet Shestakov, Alexander
Bohara, Sushil
Horváth, Samuel
Takáč, Martin
Hanzely, Slavomír
contents The Newton method is a powerful optimization algorithm, valued for its rapid local convergence and elegant geometric properties. However, its theoretical guarantees are usually limited to convex problems. In this work, we ask whether convexity is truly necessary. We introduce the concept of loss-transformation invariance, showing that damped Newton methods are unaffected by monotone transformations of the loss - apart from a simple rescaling of the step size. This insight allows difficult losses to be replaced with easier transformed versions, enabling convexification of many nonconvex problems while preserving the same sequence of iterates. Our analysis also explains the effectiveness of unconventional stepsizes in Newton's method, including values greater than one and even negative steps.
format Preprint
id arxiv_https___arxiv_org_abs_2509_25782
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Loss-Transformation Invariance in the Damped Newton Method
Shestakov, Alexander
Bohara, Sushil
Horváth, Samuel
Takáč, Martin
Hanzely, Slavomír
Optimization and Control
The Newton method is a powerful optimization algorithm, valued for its rapid local convergence and elegant geometric properties. However, its theoretical guarantees are usually limited to convex problems. In this work, we ask whether convexity is truly necessary. We introduce the concept of loss-transformation invariance, showing that damped Newton methods are unaffected by monotone transformations of the loss - apart from a simple rescaling of the step size. This insight allows difficult losses to be replaced with easier transformed versions, enabling convexification of many nonconvex problems while preserving the same sequence of iterates. Our analysis also explains the effectiveness of unconventional stepsizes in Newton's method, including values greater than one and even negative steps.
title Loss-Transformation Invariance in the Damped Newton Method
topic Optimization and Control
url https://arxiv.org/abs/2509.25782