A probabilistic generative model for semi-supervised training of coarse-grained surrogates and enforcing physical constraints through virtual observables

06/02/2020
by   Maximilian Rixner, et al.
0

The data-centric construction of inexpensive surrogates for fine-grained, physical models has been at the forefront of computational physics due to its significant utility in many-query tasks such as uncertainty quantification. Recent efforts have taken advantage of the enabling technologies from the field of machine learning (e.g. deep neural networks) in combination with simulation data. While such strategies have shown promise even in higher-dimensional problems, they generally require large amounts of training data even though the construction of surrogates is by definition a Small Data problem. Rather than employing data-based loss functions, it has been proposed to make use of the governing equations (in the simplest case at collocation points) in order to imbue domain knowledge in the training of the otherwise black-box-like interpolators. The present paper provides a flexible, probabilistic framework that accounts for physical structure and information both in the training objectives as well as in the surrogate model itself. We advocate a probabilistic (Bayesian) model in which equalities that are available from the physics (e.g. residuals, conservation laws) can be introduced as virtual observables and can provide additional information through the likelihood. We further advocate a generative model i.e. one that attempts to learn the joint density of inputs and outputs that is capable of making use of unlabeled data (i.e. only inputs) in a semi-supervised fashion in order to promote the discovery of lower-dimensional embeddings which are nevertheless predictive of the fine-grained model's output.

READ FULL TEXT

page 21

page 22

page 24

page 29

research
02/08/2021

Physics-aware, deep probabilistic modeling of multiscale dynamics in the Small Data regime

The data-based discovery of effective, coarse-grained (CG) models of hig...
research
12/30/2019

Incorporating physical constraints in a deep probabilistic machine learning framework for coarse-graining dynamical systems

Data-based discovery of effective, coarse-grained (CG) models of high-di...
research
02/11/2019

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

The automated construction of coarse-grained models represents a pivotal...
research
01/14/2021

Physics-aware, probabilistic model order reduction with guaranteed stability

Given (small amounts of) time-series' data from a high-dimensional, fine...
research
01/18/2019

Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data

Surrogate modeling and uncertainty quantification tasks for PDE systems ...
research
03/06/2017

Probabilistic Reduced-Order Modeling for Stochastic Partial Differential Equations

We discuss a Bayesian formulation to coarse-graining (CG) of PDEs where ...
research
01/22/2022

Semi-Supervised Adversarial Recognition of Refined Window Structures for Inverse Procedural Façade Modeling

Deep learning methods are notoriously data-hungry, which requires a larg...

Please sign up or login with your details

Forgot password? Click here to reset