Neural control of discrete weak formulations: Galerkin, least squares & minimal-residual methods with quasi-optimal weights

Ignacio Brevis, Ignacio Muga, Kristoffer G. van der Zee

Research output: Contribution to journalArticlepeer-review

Abstract

There is tremendous potential in using neural networks to optimize numerical methods. In this paper, we introduce and analyze a framework for the neural optimization of discrete weak formulations, suitable for finite element methods. The main idea of the framework is to include a neural-network function acting as a control variable in the weak form. Finding the neural control that (quasi-) minimizes a suitable cost (or loss) functional, then yields a numerical approximation with desirable attributes. In particular, the framework allows in a natural way the incorporation of known data of the exact solution, or the incorporation of stabilization mechanisms (e.g., to remove spurious oscillations). The main result of our analysis pertains to the well-posedness and convergence of the associated constrained-optimization problem. In particular, we prove under certain conditions, that the discrete weak forms are stable, and that quasi-minimizing neural controls exist, which converge quasi-optimally. We specialize the analysis results to Galerkin, least squares and minimal-residual formulations, where the neural-network dependence appears in the form of suitable weights. Elementary numerical experiments support our findings and demonstrate the potential of the framework.

Original languageEnglish
Article number115716
JournalComputer Methods in Applied Mechanics and Engineering
DOIs
StateAccepted/In press - 2022
Externally publishedYes

Keywords

  • Artificial neural networks
  • Data-driven discretization
  • Optimal neural control
  • Quasi-minimization
  • Quasi-optimal convergence
  • Weighted finite element methods

Fingerprint

Dive into the research topics of 'Neural control of discrete weak formulations: Galerkin, least squares & minimal-residual methods with quasi-optimal weights'. Together they form a unique fingerprint.

Cite this