activity
20242026
collaborators

6 papers

cs.AI2026

RAISE: LLM-based Automated Heuristic Design with Robust Adversary Instance Search

Fei Liu, Alessio Figalli, Patrick Owen +1

Automated Heuristic Design (AHD) with Large Language Models (LLMs) has shown remarkable progress in discovering high-quality heuristics. However, existing LLM-based AHD methods opt…

cs.LG2026

Variational inference via radial transport

Luca Ghafourpour, Sinho Chewi, Alessio Figalli +1

In variational inference (VI), the practitioner approximates a high-dimensional distribution with a simple surrogate one, often a (product) Gaussian distribution. However, in…

math.ST2025

Sharp comparisons between sliced and standard -Wasserstein distances

Guillaume Carlier, Alessio Figalli, Quentin Mérigot +1

Sliced Wasserstein distances are widely used in practice as a computationally efficient alternative to Wasserstein distances in high dimensions. In this paper, motivated by theoret…

math.PR2025

Generic regularity of equilibrium measures for the logarithmic potential with external fields

Giacomo Colombo, Alessio Figalli

It is a well-known conjecture in -models and in their discrete counterpart that, generically, external potentials should be ``off-critical'' (or, equivalently, ``regular''). Exp…

math.FA2025

Sharp Quantitative Stability for the Prékopa-Leindler and Borell-Brascamp-Lieb Inequalities

Alessio Figalli, Peter van Hintum, Marius Tiba

The Borell-Brascamp-Lieb inequality is a classical extension of the Prékopa-Leindler inequality, which in turn is a functional counterpart of the Brunn-Minkowski inequality. The s…

math.FA2024

Improved stability versions of the Prékopa-Leindler inequality

Alessio Figalli, João P. G. Ramos

We consider the problem of stability for the Prékopa-Leindler inequality. Exploiting properties of the transport map between radially decreasing functions and a suitable functiona…