We extend the well-known BFGS quasi-Newton method and its memory-limited variant LBFGS to the optimization of nonsmooth convex objectives. This is done in a rigorous fashion by ge...
We propose a convex-concave programming approach for the labeled weighted graph matching problem. The convex-concave programming formulation is obtained by rewriting the weighted ...
Mikhail Zaslavskiy, Francis Bach, Jean-Philippe Ve...
We consider a class of learning problems regularized by a structured sparsity-inducing norm defined as the sum of 2- or ∞-norms over groups of variables. Whereas much effort ha...
Image matching has been a central research topic in computer vision over the last decades. Typical approaches for correspondence involve matching features between images. In this p...
Abstract— This work considers a combination classificationregression based framework with the proposal of using learned kernels in modified support vector regression to provide...