Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 20, 2025IMA Journal of Numerical AnalysisOpen Access

Regularized black-box optimization algorithms for least-squares problems

View Full Paper
Ask AI
Bookmark
Share

Authors

YLYanjun LiuKLK. LamLRLindon Roberts

Discussion

Loading...

Member takes

Overview

The approach adapts derivative-free optimization for nonconvex functions and evaluates proximal operators.

Key Points

  • The algorithm effectively optimizes a sum of smooth and nonsmooth components with adjustments for inexact stationarity.
  • Using numerical extensions of the existing DFO-LS solver, strong practical performance is demonstrated for nonlinear least-squares.
  • The method focuses on optimizing functions without available derivatives, enhancing efficiency in practical applications.
  • Adaptations improve access to inexact stationary measures, addressing realistic challenges in optimization tasks.

Cite This Study

Liu et al. (2025) studied this question.

synapsesocial.com/papers/68f5fcce8d54a28a75cf1caahttps://doi.org/10.1093/imanum/draf093
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Derivative-free bound-constrained optimization for solving structured problems with surrogate models2024 · 1 citations
  2. 2A fresh look at nonsmooth Levenberg–Marquardt methods with applications to bilevel optimization2024 · 9 citations
  3. 3Model Construction for Convex-Constrained Derivative-Free Optimization2024
  4. 4On adaptive stochastic extended iterative methods for solving least squares2025
  5. 5Quasi-optimal least squares: inhomogeneous boundary conditions, and application with machine learning2025