Representing piecewise linear functions by functions with small arity

DOI10.1007/S00200-023-00627-1MaRDI QIDQ6962826zbMATH OpenFDO

Authors Christoph Koutschan, Bernhard A. Moser, Anton Ponomarchuk, Josef Schicho

Publication date 8 July 2025

Published in Applicable Algebra in Engineering, Communication and Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/S00200-023-00627-1

zbMATH Keywords

convex polyhedralinear combination of max-functionsMinkowski sumpiecewise linear functionReLU networkzero volume polytope

Mathematics Subject Classification ID

Real functions (26-XX) Functions of a complex variable (30-XX)

Cites work

• Canonical piecewise-linear representation
• Deep learning
• Generalization of Hinging Hyperplanes
• Max-min representation of piecewise linear functions
• Minkowski Addition of Polytopes: Computational Complexity and Applications to Gröbner Bases
• Piecewise affine functions as a difference of two convex functions
• Region configurations for realizability of lattice piecewise-linear models.
• Sharp Bounds for the Number of Regions of Maxout Networks and Vertices of Minkowski Sums

Cited in

(2)

• Absolute value linear programming
• Signed Minkowski decompositions of convex polygons into minimum simplices and factorization of max-plus functions

This page was built for publication: Representing piecewise linear functions by functions with small arity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6962826)