Discovering interpretable elastoplasticity models via the neural polynomial method enabled symbolic regressions

Abstract

Conventional neural network elastoplasticity models are often perceived as lacking interpretability. This paper introduces a two-step machine learning approach that returns mathematical models interpretable by human experts. In particular, we introduce a surrogate model where yield surfaces are expressed in terms of a set of single-variable feature mappings obtained from supervised learning. A post-processing step is then used to re-interpret the set of single-variable neural network mapping functions into mathematical form through symbolic regression. This divide-and-conquer approach provides several important advantages. First, it enables us to overcome the scaling issue of symbolic regression algorithms. From a practical perspective, it enhances the portability of learned models for partial differential equation solvers written in different programming languages. Finally, it enables us to have a concrete understanding of the attributes of the materials, such as convexity and symmetries of models, through automated derivations and reasoning. Numerical examples have been provided, along with an open-source code to enable third-party validation.

References in corpus (20)

• Intriguing properties of neural networks
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• Interpretable machine learning: definitions, methods, and applications
• Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains
• The State of Sparsity in Deep Neural Networks
• Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization
• What Does Explainable AI Really Mean? A New Conceptualization of Perspectives
• Symbolic Regression in Materials Science
• A new family of Constitutive Artificial Neural Networks towards automated model discovery
• Discovering Symbolic Models from Deep Learning with Inductive Biases
• Geometric deep learning for computational mechanics Part I: Anisotropic Hyperelasticity
• Interpretable Machine Learning for Science with PySR and SymbolicRegression.jl
• Contemporary Symbolic Regression Methods and their Relative Performance
• Meta-modeling game for deriving theoretical-consistent, micro-structural-based traction-separation laws via deep reinforcement learning
• Discovering plasticity models without stress data
• Thermodynamically Consistent Machine-Learned Internal State Variable Approach for Data-Driven Modeling of Path-Dependent Materials
• Modular machine learning-based elastoplasticity: generalization in the context of limited data
• A Greedy Search Tree Heuristic for Symbolic Regression
• A non-cooperative meta-modeling game for automated third-party calibrating, validating, and falsifying constitutive laws with parallelized adversarial attacks
• Modeling assembly bias with machine learning and symbolic regression

Cited by in corpus (2)

• Can KAN CANs? Input-convex Kolmogorov-Arnold Networks (KANs) as hyperelastic constitutive artificial neural networks (CANs)
• Explainable Hierarchical Deep Learning Neural Networks (Ex-HiDeNN)