Deep Legendre Transform

TL;DR

This paper presents a new deep learning algorithm for efficiently computing convex conjugates of differentiable convex functions, overcoming high-dimensional computational challenges and providing accurate approximations with error estimates.

Contribution

The paper introduces an implicit Fenchel formulation-based neural network method for convex conjugation, enabling efficient high-dimensional computation and exact solutions via symbolic regression.

Findings

Accurate convex conjugate computations in high dimensions

Provides a posteriori error estimates for approximations

Achieves exact solutions for specific functions using symbolic regression

Abstract

We introduce a novel deep learning algorithm for computing convex conjugates of differentiable convex functions, a fundamental operation in convex analysis with various applications in different fields such as optimization, control theory, physics and economics. While traditional numerical methods suffer from the curse of dimensionality and become computationally intractable in high dimensions, more recent neural network--based approaches scale better, but have mostly been studied with the aim of solving optimal transport problems and require the solution of complicated optimization or max--min problems. Using an implicit Fenchel formulation of convex conjugation, our approach facilitates an efficient gradient--based framework for the minimization of approximation errors and, as a byproduct, also provides a posteriori estimates of the approximation accuracy. Numerical experiments demonstrate our method's ability to deliver accurate results across different high-dimensional examples. Moreover, by employing symbolic regression with Kolmogorov--Arnold networks, it is able to obtain the exact convex conjugates of specific convex functions.