A physics-aware, probabilistic machine learning framework for coarse-graining high-dimensional systems in the Small Data regime

Abstract

The automated construction of coarse-grained models represents a pivotal component in computer simulation of physical systems and is a key enabler in various analysis and design tasks related to uncertainty quantification. Pertinent methods are severely inhibited by the high-dimension of the parametric input and the limited number of training input/output pairs that can be generated when computationally demanding forward models are considered. Such cases are frequently encountered in the modeling of random heterogeneous media where the scale of the microstructure necessitates the use of high-dimensional random vectors and very fine discretizations of the governing equations. The present paper proposes a probabilistic Machine Learning framework that is capable of operating in the presence of Small Data by exploiting aspects of the physical structure of the problem as well as contextual knowledge. As a result, it can perform comparably well under extrapolative conditions. It unifies the tasks of dimensionality and model-order reduction through an encoder-decoder scheme that simultaneously identifies a sparse set of salient lower-dimensional microstructural features and calibrates an inexpensive, coarse-grained model which is predictive of the output. Information loss is accounted for and quantified in the form of probabilistic predictive estimates. The learning engine is based on Stochastic Variational Inference. We demonstrate how the variational objectives can be used not only to train the coarse-grained model, but also to suggest refinements that lead to improved predictions.

arXiv comment43 pages, 17 figures

References in corpus (25)

• Auto-Encoding Variational Bayes
• Variational Inference: A Review for Statisticians
• The information bottleneck method
• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Tutorial on Variational Autoencoders
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
• Active subspace methods in theory and practice: applications to kriging surfaces
• Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification
• Deep convolutional encoder-decoder networks for uncertainty quantification of dynamic multiphase flow in heterogeneous media
• Variational Relevance Vector Machines
• Inferring solutions of differential equations using noisy multi-fidelity data
• Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial Differential Equations
• Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence (2000)
• Data-Driven Model Reduction for the Bayesian Solution of Inverse Problems
• Variational Bayesian Inference with Stochastic Search
• Structure Preserving Model Reduction of Parametric Hamiltonian Systems
• Physics-Informed CoKriging: A Gaussian-Process-Regression-Based Multifidelity Method for Data-Model Convergence
• Structured Bayesian Gaussian process latent variable model: applications to data-driven dimensionality reduction and high-dimensional inversion
• Predictive Coarse-Graining
• Linking Gaussian Process regression with data-driven manifold embeddings for nonlinear data fusion
• Bayesian model and dimension reduction for uncertainty propagation: applications in random media
• Basis adaptation and domain decomposition for steady partial differential equations with random coefficients
• Probabilistic Reduced-Order Modeling for Stochastic Partial Differential Equations

Cited by in corpus (8)

• Uncertainty Quantification in Scientific Machine Learning: Methods, Metrics, and Comparisons
• Solving inverse-PDE problems with physics-aware neural networks
• Symplectic Neural Networks in Taylor Series Form for Hamiltonian Systems
• Incorporating physical constraints in a deep probabilistic machine learning framework for coarse-graining dynamical systems
• A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables
• Physics-Aware Neural Implicit Solvers for multiscale, parametric PDEs with applications in heterogeneous media
• Validation and parameter optimization of a hybrid embedded/homogenized solid tumor perfusion model
• Physics-aware, probabilistic model order reduction with guaranteed stability