English

A linear-time algorithm to compute the conjugate of nonconvex bivariate piecewise linear-quadratic functions

Optimization and Control 2025-05-13 v1 Symbolic Computation

Abstract

We propose the first linear-time algorithm to compute the conjugate of (nonconvex) bivariate piecewise linear-quadratic (PLQ) functions (bivariate quadratic functions defined on a polyhedral subdivision). Our algorithm starts with computing the convex envelope of each quadratic piece obtaining rational functions (quadratic over linear) defined over a polyhedral subdivision. Then we compute the conjugate of each resulting piece to obtain piecewise quadratic functions defined over a parabolic subdivision. Finally we compute the maximum of all those functions to obtain the conjugate as a piecewise quadratic function defined on a parabolic subdivision. The resulting algorithm runs in linear time if the initial subdivision is a triangulation (or has a uniform upper bound on the number of vertexes for each piece). Our open-source implementation in MATLAB uses symbolic computation and rational numbers to avoid floating-point errors, and merges pieces as soon as possible to minimize computation time.

Keywords

Cite

@article{arxiv.2505.06442,
  title  = {A linear-time algorithm to compute the conjugate of nonconvex bivariate piecewise linear-quadratic functions},
  author = {Tanmaya Karmarkar and Yves Lucet},
  journal= {arXiv preprint arXiv:2505.06442},
  year   = {2025}
}

Comments

to be published in Journal of Global Optimization

R2 v1 2026-06-28T23:27:51.279Z