A practical randomized trust-region method to escape saddle points in high dimension

Documenti analoghi

• An adaptive framework for first-order gradient methods
  di: Hu, Xiaozhe, et al.
  Pubblicazione: (2026)
• Fast Reflected Forward-Backward algorithm: achieving fast convergence rates for convex optimization with linear cone constraints
  di: Bot, Radu Ioan, et al.
  Pubblicazione: (2024)
• Restarts subject to approximate sharpness: A parameter-free and optimal scheme for first-order methods
  di: Adcock, Ben, et al.
  Pubblicazione: (2023)
• Optimization in Theory and Practice
  di: Wright, Stephen J.
  Pubblicazione: (2025)
• The rate of convergence of Bregman proximal methods: Local geometry vs. regularity vs. sharpness
  di: Azizian, Waïss, et al.
  Pubblicazione: (2022)