Single- to multi-fidelity history-dependent learning with uncertainty quantification and disentanglement: application to data-driven constitutive modeling
arXiv:2507.13416 · doi:10.1016/j.cma.2025.118479
Abstract
Data-driven learning is generalized to consider history-dependent multi-fidelity data, while quantifying epistemic uncertainty and disentangling it from data noise (aleatoric uncertainty). This generalization is hierarchical and adapts to different learning scenarios: from training the simplest single-fidelity deterministic neural networks up to the proposed multi-fidelity variance estimation Bayesian recurrent neural networks. The versatility and generality of the proposed methodology are demonstrated by applying it to different data-driven constitutive modeling scenarios that include multiple fidelities with and without aleatoric uncertainty (noise). The method accurately predicts the response and quantifies model error while also discovering the noise distribution (when present). This opens opportunities for future real-world applications in diverse scientific and engineering domains; especially, the most challenging cases involving design and analysis under uncertainty.
40 pages, 32 figures
References in corpus (19)
- Adam: A Method for Stochastic Optimization
- Variational Inference: A Review for Statisticians
- MCMC using Hamiltonian dynamics
- A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
- Proceedings of the 29th International Conference on Machine Learning (ICML-12)
- Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
- A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems
- Uncertainty Quantification in Scientific Machine Learning: Methods, Metrics, and Comparisons
- Multi-fidelity Bayesian Neural Networks: Algorithms and Applications
- Bayesian Recurrent Neural Networks
- Multi-fidelity surrogate modeling using long short-term memory networks
- Thermodynamically Consistent Machine-Learned Internal State Variable Approach for Data-Driven Modeling of Path-Dependent Materials
- How to Evaluate Uncertainty Estimates in Machine Learning for Regression?
- Adaptivity for clustering-based reduced-order modeling of localized history-dependent phenomena
- Consistent machine learning for topology optimization with microstructure-dependent neural network material models
- Multi-Fidelity Bayesian Optimization via Deep Neural Networks
- Cooperative data-driven modeling
- Are you using test log-likelihood correctly?
- Cooperative Variance Estimation and Bayesian Neural Networks for Disentangling Aleatoric and Epistemic Uncertainties