Input convex neural networks: universal approximation theorem and implementation for isotropic polyconvex hyperelastic energies

TL;DR

This paper introduces a neural network framework for isotropic hyperelasticity that enforces physical constraints, proves a universal approximation theorem, and effectively models complex energy functions.

Contribution

It presents a novel input convex neural network architecture with a proven universal approximation theorem for isotropic polyconvex energies.

Findings

The network can approximate any frame-indifferent, isotropic polyconvex energy.

It effectively captures non-polyconvex energies and computes polyconvex hulls.

The approach outperforms existing methods in approximation accuracy.

Abstract

This paper presents a novel framework of neural networks for isotropic hyperelasticity that enforces necessary physical and mathematical constraints while simultaneously satisfying the universal approximation theorem. The two key ingredients are an input convex network architecture and a formulation in the elementary polynomials of the signed singular values of the deformation gradient. In line with previously published networks, it can rigorously capture frame-indifference and polyconvexity - as well as further constraints like balance of angular momentum and growth conditions. However and in contrast to previous networks, a universal approximation theorem for the proposed approach is proven. To be more explicit, the proposed network can approximate any frame-indifferent, isotropic polyconvex energy (provided the network is large enough). This is possible by working with a sufficient and necessary criterion for frame-indifferent, isotropic polyconvex functions. Comparative studies with existing approaches identify the advantages of the proposed method, particularly in approximating non-polyconvex energies as well as computing polyconvex hulls.