The Multi-Block DC Function Class: Theory, Algorithms, and Applications

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Fatemi, Pouria, Maskan, Hoomaan, Yurtsever, Alp, Sra, Suvrit
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910146453569536
author Fatemi, Pouria
Maskan, Hoomaan
Yurtsever, Alp
Sra, Suvrit
author_facet Fatemi, Pouria
Maskan, Hoomaan
Yurtsever, Alp
Sra, Suvrit
contents We present the Multi-Block DC (BDC) class, a rich class of structured nonconvex functions that admit a DC ("difference-of-convex") decomposition across parameter blocks. This multi-block class not only subsumes the usual DC programming, but also turns out to be provably more powerful. Specifically, we demonstrate how standard models (e.g., polynomials and tensor factorization) must have DC decompositions of exponential size, while their BDC formulation is polynomial. This separation in complexity also underscores another key aspect: unlike DC formulations, obtaining BDC formulations for problems is vastly easier and constructive. We illustrate this aspect by presenting explicit BDC formulations for modern tasks such as deep ReLU networks, a result with no known equivalent in the DC class. Moreover, we complement the theory by developing algorithms with non-asymptotic convergence theory, including both batch and stochastic settings, and demonstrate the broad applicability of our method through several applications.
format Preprint
id arxiv_https___arxiv_org_abs_2604_17560
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Multi-Block DC Function Class: Theory, Algorithms, and Applications
Fatemi, Pouria
Maskan, Hoomaan
Yurtsever, Alp
Sra, Suvrit
Optimization and Control
90C26
We present the Multi-Block DC (BDC) class, a rich class of structured nonconvex functions that admit a DC ("difference-of-convex") decomposition across parameter blocks. This multi-block class not only subsumes the usual DC programming, but also turns out to be provably more powerful. Specifically, we demonstrate how standard models (e.g., polynomials and tensor factorization) must have DC decompositions of exponential size, while their BDC formulation is polynomial. This separation in complexity also underscores another key aspect: unlike DC formulations, obtaining BDC formulations for problems is vastly easier and constructive. We illustrate this aspect by presenting explicit BDC formulations for modern tasks such as deep ReLU networks, a result with no known equivalent in the DC class. Moreover, we complement the theory by developing algorithms with non-asymptotic convergence theory, including both batch and stochastic settings, and demonstrate the broad applicability of our method through several applications.
title The Multi-Block DC Function Class: Theory, Algorithms, and Applications
topic Optimization and Control
90C26
url https://arxiv.org/abs/2604.17560