Any-stepsize Gradient Descent for Separable Data under Fenchel–Young Losses

NeurIPSSpotlight2025

Authors: Han Bao, Shinsaku Sakaue, Yuki Takezawa
Venue: NeurIPS 2025
Track: Spotlight

TL;DR

The gradient descent (GD) has been one of the most common optimizer in machine learning…

Opening excerpt from the authors’ abstract. source

Read the paper

OpenReview

← All NeurIPS 2025 Spotlight papers · Browse the whole archive