Id: arxiv/2608.01243 Url: https://arxiv.org/abs/2608.01243 Tags: math.OC, math.NA Score: 0 Title: Revisiting Decomposition-Invariant Conditional Gradient Methods for Polytopes Domain: math.OC Source: arxiv Authors: Dan Garber Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-08-02 Page: https://delphi.science/r/arxiv/2608.01243 We revisit Decomposition-Invariant Conditional Gradient methods, originally introduced by Garber and Meshi in 2016, for minimizing a convex and $β$-smooth function over a polytope in $\reals^n$, under an $α$-quadratic growth condition. For 2-level polytopes we design a simple and parameter-free dyadic step-size rule that yields a linear convergence rate which scales with the dimension of the optimal face and not with the ambient dimension as in standard away-step-based conditional gradient methods for polytopes. For general polytopes, under a slightly stronger condition of…