Id: arxiv/2603.07075 Url: https://arxiv.org/abs/2603.07075 Tags: math.OC, math.NA Score: 0 Title: A unified high-resolution ODE framework for first-order methods Domain: math.OC Source: arxiv Authors: Lixia Wang, Hao Luo Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-03-07 Page: https://delphi.science/r/arxiv/2603.07075 For a generic discrete-time algorithm (DTA): $z^+=g(z,s)$, where $s$ is the step size, Lu (Math. Program., 194(1):1061--1112, 2022) proposed an $O(s^r)$-resolution ordinary differential equation (ODE) framework based on the backward error analysis, which can be used to analyze many DTAs satisfying the fixed point assumption $g(z,0)=z$ such as gradient descent, extra gradient method and primal-dual hybrid gradient (PDHG). However, most first-order methods with momentum violate this critical assumption. To address this issue, in this work, we introduce a novel $O((\sqrt{s})^r)$-resolution ODE…