Mutual Information for Explainable Deep Learning of Multiscale Systems
arXiv:2009.04570 · doi:10.1016/j.jcp.2021.110551
Abstract
Timely completion of design cycles for complex systems ranging from consumer electronics to hypersonic vehicles relies on rapid simulation-based prototyping. The latter typically involves high-dimensional spaces of possibly correlated control variables (CVs) and quantities of interest (QoIs) with non-Gaussian and possibly multimodal distributions. We develop a model-agnostic, moment-independent global sensitivity analysis (GSA) that relies on differential mutual information to rank the effects of CVs on QoIs. The data requirements of this information-theoretic approach to GSA are met by replacing computationally intensive components of the physics-based model with a deep neural network surrogate. Subsequently, the GSA is used to explain the network predictions, and the surrogate is deployed to close design loops. Viewed as an uncertainty quantification method for interrogating the surrogate, this framework is compatible with a wide variety of black-box models. We demonstrate that the surrogate-driven mutual information GSA provides useful and distinguishable rankings on two applications of interest in energy storage. Consequently, our information-theoretic GSA provides an "outer loop" for accelerated product design by identifying the most and least sensitive input directions and performing subsequent optimization over appropriately reduced parameter subspaces.
arXiv comment27 pages, 8 figures. Added additional examples
References in corpus (19)
- Estimating Mutual Information
- Learning Important Features Through Propagating Activation Differences
- DGM: A deep learning algorithm for solving partial differential equations
- Kernel density estimation via diffusion
- Surrogate Modeling for Fluid Flows Based on Physics-Constrained Deep Learning Without Simulation Data
- Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
- A composite neural network that learns from multi-fidelity data: Application to function approximation and inverse PDE problems
- Equitability, mutual information, and the maximal information coefficient
- A unified deep artificial neural network approach to partial differential equations in complex geometries
- Sanity Checks for Saliency Maps
- Quantifying total uncertainty in physics-informed neural networks for solving forward and inverse stochastic problems
- Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification
- Adversarial Uncertainty Quantification in Physics-Informed Neural Networks
- Sensitivity, robustness and identifiability in stochastic chemical kinetics models
- Gradients of Counterfactuals
- A Relative Entropy Rate Method for Path Space Sensitivity Analysis of Stationary Complex Stochastic Dynamics
- GINNs: Graph-Informed Neural Networks for Multiscale Physics
- Causality and Bayesian network PDEs for multiscale representations of porous media
- Robust information divergences for model-form uncertainty arising from sparse data in random PDE