Physics-informed deep generative models

Abstract

We consider the application of deep generative models in propagating uncertainty through complex physical systems. Specifically, we put forth an implicit variational inference formulation that constrains the generative model output to satisfy given physical laws expressed by partial differential equations. Such physics-informed constraints provide a regularization mechanism for effectively training deep probabilistic models for modeling physical systems in which the cost of data acquisition is high and training data-sets are typically small. This provides a scalable framework for characterizing uncertainty in the outputs of physical systems due to randomness in their inputs or noise in their observations. We demonstrate the effectiveness of our approach through a canonical example in transport dynamics.

arXiv commentAccepted by the NIPS workshop 2018 of Bayesian Deep Learning. arXiv admin note: text overlap with arXiv:1811.04026

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations
• Physics Informed Deep Learning (Part I): Data-driven Solutions of Nonlinear Partial Differential Equations
• Deep Hidden Physics Models: Deep Learning of Nonlinear Partial Differential Equations
• Bayesian Compression for Deep Learning
• Deep Gaussian Processes for Regression using Approximate Expectation Propagation
• Fixing a Broken ELBO
• Variational Implicit Processes

Cited by in corpus (22)

• Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systems
• Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks
• A Review of Machine Learning Methods Applied to Structural Dynamics and Vibroacoustic
• When and why PINNs fail to train: A neural tangent kernel perspective
• Deep learning of free boundary and Stefan problems
• Physics-informed Autoencoders for Lyapunov-stable Fluid Flow Prediction
• A neural network multigrid solver for the Navier-Stokes equations
• Scalable Uncertainty Quantification for Deep Operator Networks using Randomized Priors
• A comparative study of physics-informed neural network models for learning unknown dynamics and constitutive relations
• Energy-based error bound of physics-informed neural network solutions in elasticity
• Solving Bayesian Inverse Problems via Variational Autoencoders
• Physics Informed Deep Learning for Transport in Porous Media. Buckley Leverett Problem
• Data-driven Identification of 2D Partial Differential Equations using extracted physical features
• Variational Autoencoders for Learning Nonlinear Dynamics of Physical Systems
• POD-DL-ROM: enhancing deep learning-based reduced order models for nonlinear parametrized PDEs by proper orthogonal decomposition
• Deep Learning of Dynamic Subsurface Flow via Theory-guided Generative Adversarial Network
• Structure Preservation for the Deep Neural Network Multigrid Solver
• Physics-Integrated Variational Autoencoders for Robust and Interpretable Generative Modeling
• Exact artificial boundary conditions of 1D semi-discretized peridynamics
• BINet: Learning to Solve Partial Differential Equations with Boundary Integral Networks
• AdvNF: Reducing Mode Collapse in Conditional Normalising Flows using Adversarial Learning
• Learning To Solve Differential Equations Across Initial Conditions