AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Newton Method Revisited: Global Convergence Rates up to $O(1/k^3)$ for Stepsize Schedules and Linesearch Procedures

Slavomir Hanzely, Farshed Abdukhakimov, Martin Takáč

ICLR 2026 · Optimization › Nonconvex and smooth optimization · approach: Adaptive first-order convex methods · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Root Newton stepsize instead of AICNRoot Newton · Logistic regression and polytope feasibilityconvergence speedNo clear effect
use the universal Newton backtracking method instead of Super-universal NewtonUniversal Newton · Logistic regression and polytope feasibilityconvergence speedImproved
vary the exponent beta of the regularization term (ablation)Universal Newton and Super-universal Newton · Logistic regression and polytope feasibilityconvergence speedNo clear effect
use GRLS or Greedy Newton linesearch instead of Armijo or fixed-stepsize NewtonNewton method · Logistic regression and polytope feasibilityconvergence speedImproved
use GRLS linesearch instead of other linesearchesNewton method · Rosenbrock function (nonconvex, 5 random initializations)convergenceImproved
use a first-order gradient method instead of Newton-type methodsGradient Method · Logistic regression and polytope feasibilityper-iteration convergenceHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy