A Regularized Hessian-Free Inexact Newton-Type Method with Global $\mathcal{O}(k^{-2})$ Convergence

Leandro Farias Maia; Antonio Victor B. Nascimento; Paulo Sergio M. Santos; Gilson N. Silva

A Regularized Hessian-Free Inexact Newton-Type Method with Global $\mathcal{O}(k^{-2})$ Convergence

Leandro Farias Maia, Antonio Victor B. Nascimento, Paulo Sergio M. Santos, Gilson N. Silva

arXivarxiv/2604.27406Apr 30, 2026math.OCscore 0

We propose a regularized Hessian-free Newton-type method for minimizing smooth convex functions with Lipschitz continuous Hessians. The algorithm constructs an approximate Hessian by finite differences and selects the regularization parameter through an adaptive criterion that ensures sufficient decrease and gradient control. We prove that the method achieves a global $\mathcal{O}(k^{-2})$ convergence rate, matching the best known bound for second-order methods. A modified variant incorporating the exact Hessian when available enjoys local quadratic convergence under standard assumptions.…

Source JSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2604.27406
curl -X POST /api/v1/records/arxiv/2604.27406/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2604.27406/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'