Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 20, 2025Open Access

Modified Block Newton Algorithm for ₀- Regularized Optimization

View Full Paper
Ask AI
Bookmark
Share

Authors

YYYangyi YeQLQingna Li

Discussion

Loading...

Member takes

Overview

This approach improves global convergence in sparse optimization, suggesting enhanced efficiency with regularization.

Key Points

  • The globally convergent method achieves efficiency in sparse optimization by optimizing the block diagonal of the Jacobian.
  • Numerical results validate that the proposed algorithm maintains local quadratic convergence while successfully addressing matrix singularity.
  • Incorporating a line search strategy allows for the rigorous global convergence of the modified Newton algorithm.
  • The approach minimizes computational burden without sacrificing performance across various sparse optimization scenarios.

Cite This Study

Ye et al. (2025) studied this question.

synapsesocial.com/papers/68f5fcdc8d54a28a75cf256chttps://doi.org/10.48550/arxiv.2507.03566
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. 1Reduced-Space Iteratively Reweighted Second-Order Methods for Nonconvex Sparse Regularization2024
  2. 2Block cubic Newton with greedy selection2024
  3. 3Coderivative-Based Newton Methods in Structured Nonconvex and Nonsmooth Optimization2024
  4. 4Online learning guided quasi-Newton methods with global non-asymptotic convergence2026 · 1 citations
  5. 5Multilevel Regularized Newton Methods with Fast Convergence Rates2024