Scalable DC Optimization via Adaptive Frank-Wolfe Algorithms
Reviewed by Pithpith:2FAL7C5Popen to challenge →
read the original abstract
We consider the problem of minimizing a difference of (smooth) convex functions over a compact convex feasible region $P$, i.e., $\min_{x \in P} f(x) - g(x)$, with smooth $f$ and Lipschitz continuous $g$. This computational study builds upon and complements the framework of Maskan et al. [2025] by integrating advanced Frank-Wolfe variants to reduce computational overhead. We empirically show that constrained DC problems can be efficiently solved using a combination of the Blended Pairwise Conditional Gradients (BPCG) algorithm [Tsuji et al., 2022] with warm-starting and the adaptive error bound from Maskan et al. [2025]. The result is a highly efficient and scalable projection-free algorithm for constrained DC optimization.
This paper has not been read by Pith yet.
Forward citations
Cited by 2 Pith papers
-
Analytical Evaluation of DCA Convergence Properties for Minimizing Prediction Functions of Gaussian RBF Support Vector Regression
Derives closed-form μ and L bounds for DCA on RBF-SVR prediction minimization, showing C_α ρ as the primary quantity characterizing convergence and initial-point dependence.
-
Graph Isomorphism: Mixed-Integer Convex Optimization from First-Order Methods
A mixed-integer convex formulation for graph isomorphism is solved via first-order methods and variable fixing, outperforming integer feasibility on 6 of 12 graph families and matching on symmetric ones.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.