Nonmonotone Coderivative-Based Newton Methods for Nonsmooth Optimization with Machine Learning Applications
Abstract
Newton-type algorithms are among the most effective methods for solving optimization problems because of their rapid local convergence. However, extending these methods to nonsmooth optimization is challenging due to the difficulty of incorporating second-order information. To address this issue, we propose a coderivative-based Newton framework for solving both unconstrained and constrained nonsmooth optimization problems using tools from variational analysis and generalized differentiation. The proposed method employs generalized Hessians, defined as coderivatives of the subgradient mapping, and is applicable to both C 1,1 functions and convex composite optimization problems with extended-real-valued components. To enhance robustness, the algorithm incorporates an adaptive regularization strategy that enables it to handle problems whose generalized Hessians are positive semidefinite. In addition, a Hybrid Adaptive Nonmonotone (HAN) line search scheme is developed as a globalization technique to improve global convergence and practical performance. Under standard assumptions, both the exact and inexact versions of the algorithm are globally convergent and achieve local superlinear convergence when the associated subgradient mapping satisfies the semismooth* property. The proposed framework is further extended to convex composite optimization through the forward-backward envelope, allowing it to handle problems with or without strong convexity in the smooth component of the objective function. Numerical experiments on Lasso, logistic Lasso, and support vector machine (SVM) problems demonstrate the efficiency and robustness of the proposed algorithm. Comparisons with several well-established first-order and second-order methods for nonsmooth optimization show that the proposed approach is computationally competitive while maintaining strong convergence properties.
Related
- A Proximal Gradient Method for Nonconvex Nonsmooth Optimization with Applications
- Two applications of the point-free coderivative
- Exact Credibilistic Expectations of Nonmonotone Fuzzy Transformations with Applications to Higher-Order Moments
- Linear convergence rate analysis of Proximal-Tracking for nonsmooth distributed optimization
- Exact penalty methods with adaptive discretization for nonsmooth semi-infinite optimization
BibTeX
@misc{kazemdehbashi2026nonmonotone,
title = {{Nonmonotone Coderivative-Based Newton Methods for Nonsmooth Optimization with Machine Learning Applications}},
author = {Sina Kazemdehbashi and Yanchao Liu and Boris S. Mordukhovich},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.09364},
eprint = {2610.09364},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.09364},
}
RIS
TY - UNPB
TI - Nonmonotone Coderivative-Based Newton Methods for Nonsmooth Optimization with Machine Learning Applications
AU - Sina Kazemdehbashi
AU - Yanchao Liu
AU - Boris S. Mordukhovich
PY - 2026
DA - 2026/10/07
PB - arXiv
DO - 10.48550/arxiv.2610.09364
UR - https://arxiv.org/abs/2610.09364
AB - Newton-type algorithms are among the most effective methods for solving optimization problems because of their rapid local convergence. However, extending these methods to nonsmooth optimization is challenging due to the difficulty of incorporating second-order information. To address this issue, we propose a coderivative-based Newton framework for solving both unconstrained and constrained nonsmooth optimization problems using tools from variational analysis and generalized differentiation. The proposed method employs generalized Hessians, defined as coderivatives of the subgradient mapping, and is applicable to both C ^1,1 functions and convex composite optimization problems with extended-real-valued components. To enhance robustness, the algorithm incorporates an adaptive regularization strategy that enables it to handle problems whose generalized Hessians are positive semidefinite. In addition, a Hybrid Adaptive Nonmonotone (HAN) line search scheme is developed as a globalization technique to improve global convergence and practical performance. Under standard assumptions, both the exact and inexact versions of the algorithm are globally convergent and achieve local superlinear convergence when the associated subgradient mapping satisfies the semismooth* property. The proposed framework is further extended to convex composite optimization through the forward-backward envelope, allowing it to handle problems with or without strong convexity in the smooth component of the objective function. Numerical experiments on Lasso, logistic Lasso, and support vector machine (SVM) problems demonstrate the efficiency and robustness of the proposed algorithm. Comparisons with several well-established first-order and second-order methods for nonsmooth optimization show that the proposed approach is computationally competitive while maintaining strong convergence properties.
ER -