Extreme sparsification of physics-augmented neural networks for interpretable model discovery in mechanics

DOI10.1016/J.CMA.2024.116973zbMATH Open1539.74544MaRDI QIDQ6550128FDOQ6550128

Authors: Jan N. Fuhg, Reese Edward Jones, Nikolaos Bouklas 

Publication date: 4 June 2024

Published in: Computer Methods in Applied Mechanics and Engineering (Search for Journal in Brave)

Recommendations

• Physics-constrained symbolic model discovery for polyconvex incompressible hyperelastic materials
• A mechanics‐informed artificial neural network approach in data‐driven constitutive modeling
• On physics-informed data-driven isotropic and anisotropic constitutive models through probabilistic machine learning and space-filling sampling
• A new family of constitutive artificial neural networks towards automated model discovery

zbMATH Keywords

solid mechanicsdata-driven constitutive modelsphysics-augmented machine learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Numerical and other methods in solid mechanics (74S99)

Cites Work

• SINDy-PI: a robust algorithm for parallel implicit sparse identification of nonlinear dynamics
• Discovering governing equations from data by sparse identification of nonlinear dynamical systems
• Convexity conditions and existence theorems in nonlinear elasticity
• An Introduction to Numerical Analysis
• Large deformation isotropic elasticity – on the correlation of theory and experiment for incompressible rubberlike solids
• Invariant formulation of hyperelastic transverse isotropy based on polyconvex free energy functions.
• Title not available (Why is that?)
• Fitting hyperelastic models to experimental data
• Orthotropic yield criterion for hexagonal closed packed metals
• Title not available (Why is that?)
• Parameter estimation of hyperelasticity relations of generalized polynomial-type with constraint conditions
• Convex Entropies and Hyperbolicity for General Euler Equations
• Title not available (Why is that?)
• Title not available (Why is that?)
• Matrixungleichungen und die Konvexität der Fließfläche
• A new family of constitutive artificial neural networks towards automated model discovery
• Mechanik der festen Körper im plastisch-deformablen Zustand.
• Sobolev training of thermodynamic-informed neural networks for interpretable elasto-plasticity models with level set hardening
• Local approximate Gaussian process regression for data-driven constitutive models: development and comparison with neural networks
• A machine learning based plasticity model using proper orthogonal decomposition
• Unsupervised discovery of interpretable hyperelastic constitutive laws
• Finite electro-elasticity with physics-augmented neural networks
• Automated constitutive modeling of isotropic hyperelasticity based on artificial neural networks
• On physics-informed data-driven isotropic and anisotropic constitutive models through probabilistic machine learning and space-filling sampling
• Machine learning constitutive models of elastomeric foams
• Data-driven anisotropic finite viscoelasticity using neural ordinary differential equations
• A theory of large elastic deformation.
• Automated discovery of generalized standard material models with EUCLID
• Enhancing phenomenological yield functions with data: challenges and opportunities
• Modular machine learning-based elastoplasticity: generalization in the context of limited data
• Neural network-based multiscale modeling of finite strain magneto-elasticity with relaxed convexity criteria

Cited In (5)

• On sparse regression, \(L_p\)-regularization, and automated model discovery
• Physics-constrained symbolic model discovery for polyconvex incompressible hyperelastic materials
• Improving the performance of Stein variational inference through extreme sparsification of physically-constrained neural network models
• Discovering uncertainty: Bayesian constitutive artificial neural networks
• Discovering interpretable elastoplasticity models via the neural polynomial method enabled symbolic regressions

This page was built for publication: Extreme sparsification of physics-augmented neural networks for interpretable model discovery in mechanics

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6550128)