arXiv subjects

Seta Rakotomandimby

Publications and source records attributed to Seta Rakotomandimby.

3 recordsLinked to original sources

arXiv2026

Efficient Quadratic Corrections for Frank-Wolfe Algorithms

We develop a Frank-Wolfe algorithm with corrective steps, generalizing previous algorithms including blended conditional gradients, blended pairwise conditional gradients, and fully-corrective Frank-Wolfe. For this, we prove tight convergence guarantees together with an optimal face identification property. Furthermore, we propose two highly efficient corrective steps for convex quadratic objectives based on linear optimization or linear system solving, akin to Wolfe's minimum-norm point, and show that they converge in finite time under suitable conditions. Beyond optimization problems that are directly quadratic, we revisit two algorithms - split conditional gradient and second-order conditional gradient sliding - which can leverage quadratic corrections to accelerate their quadratic subproblems. We demonstrate improved convergence rates for the first and broader applicability for the second, which may be of independent interest. Finally, we show substantial computational speedups for Frank-Wolfe-based algorithms with quadratic corrections across the considered problem classes.

math.OC↗

arXiv2026

What are Capra-Convex Sets?

This paper focuses on a specific form of abstract convexity known as Capra-convexity, where a constant along primal rays (Capra) coupling replaces the scalar product used in standard convex analysis to define generalized Fenchel conjugacies. A key motivating result is that the ${\ell}$0 pseudonorm - which counts the number of nonzero components in a vector - is equal to its Capra-biconjugate. This implies that ${\ell}$0 is a Capra-convex function, highlighting potential applications in statistics and machine learning, particularly for enforcing sparsity in models. Building on prior work characterizing the Capra-subdifferential of ${\ell}$0 and the role of source norms in defining the Capra-coupling, the paper provides a characterization of Capra-convex sets.

math.OC↗

arXiv2024

Learning with Fitzpatrick Losses

Fenchel-Young losses are a family of convex loss functions, encompassing the squared, logistic and sparsemax losses, among others. Each Fenchel-Young loss is implicitly associated with a link function, for mapping model outputs to predictions. For instance, the logistic loss is associated with the soft argmax link function. Can we build new loss functions associated with the same link function as Fenchel-Young losses? In this paper, we introduce Fitzpatrick losses, a new family of convex loss functions based on the Fitzpatrick function. A well-known theoretical tool in maximal monotone operator theory, the Fitzpatrick function naturally leads to a refined Fenchel-Young inequality, making Fitzpatrick losses tighter than Fenchel-Young losses, while maintaining the same link function for prediction. As an example, we introduce the Fitzpatrick logistic loss and the Fitzpatrick sparsemax loss, counterparts of the logistic and the sparsemax losses. This yields two new tighter losses associated with the soft argmax and the sparse argmax, two of the most ubiquitous output layers used in machine learning. We study in details the properties of Fitzpatrick losses and in particular, we show that they can be seen as Fenchel-Young losses using a modified, target-dependent generating function. We demonstrate the effectiveness of Fitzpatrick losses for label proportion estimation.

stat.ML↗