CCCP Is Frank-Wolfe in Disguise
Abstract
This paper uncovers a simple but rather surprising connection: it shows that the well-known convex-concave procedure (CCCP) and its generalization to constrained problems are both special cases of the Frank-Wolfe (FW) method. This connection not only provides insight of deep (in our opinion) pedagogical value, but also transfers the recently discovered convergence theory of nonconvex Frank-Wolfe methods immediately to CCCP, closing a long-standing gap in its non-asymptotic convergence theory. We hope the viewpoint uncovered by this paper spurs the transfer of other advances made for FW to both CCCP and its generalizations.
Cite
Text
Yurtsever and Sra. "CCCP Is Frank-Wolfe in Disguise." Neural Information Processing Systems, 2022.Markdown
[Yurtsever and Sra. "CCCP Is Frank-Wolfe in Disguise." Neural Information Processing Systems, 2022.](https://mlanthology.org/neurips/2022/yurtsever2022neurips-cccp/)BibTeX
@inproceedings{yurtsever2022neurips-cccp,
title = {{CCCP Is Frank-Wolfe in Disguise}},
author = {Yurtsever, Alp and Sra, Suvrit},
booktitle = {Neural Information Processing Systems},
year = {2022},
url = {https://mlanthology.org/neurips/2022/yurtsever2022neurips-cccp/}
}