Parameter inference with deep jointly informed neural networks

Correspondence *K. D. Humbird, Lawrence Livermore National Laboratory, 7000 East Ave, Livermore, CA 94550, Email: humbird1@llnl.gov Summary A common challenge in modeling inertial confinement fusion (ICF) experiments with computer simulations is that many of the simulation inputs are unknown and cannot be directly measured. Often, parameters that are measured in the experiment are used to infer the unknown inputs by solving the inverse problem: finding the set of simulation inputs that result in outputs consistent with the experimental observations. In inertial confinement fusion, this process is often referred to as a “post-shot analysis”. Post-shot analyses are challenging as the inverse problem is often highly degenerate, the input parameter space is vast, and simulations are computationally expensive. In this work, deep neural network models equipped with model uncertainty estimates are used to train inverse models, which map directly from output to input space, to find the distribution of post-shot simulations that are consistent with experimental observations. The inverse model approach is compared to Markov chain Monte Carlo (MCMC) sampling of the forward model, which maps from input to output space, for parameter inference tasks of varying complexity. The inverse models perform best when searching vast parameter spaces for post-shot simulations that are consistent with a large number of observables, where MCMC sampling can be prohibitively expensive. We demonstrate how augmenting inverse models with autoencoders enable the inclusion of several dozen observables in the inverse mapping, reducing the degeneracy of the model and improving the accuracy of the post-shot analysis.

[1]  Michael J. Watts,et al.  IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS Publication Information , 2020, IEEE Transactions on Neural Networks and Learning Systems.

[2]  Yarin Gal,et al.  Uncertainty in Deep Learning , 2016 .

[3]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[4]  D S Clark,et al.  X-ray shadow imprint of hydrodynamic instabilities on the surface of inertial confinement fusion capsules by the fuel fill tube. , 2017, Physical review. E.

[5]  David G. T. Denison,et al.  Bayesian MARS , 1998, Stat. Comput..

[6]  Niels Bohr,et al.  INERTIAL CONFINEMENT FUSION , 2006 .

[7]  Jim Gaffney,et al.  Using deep neural networks to augment NIF post-shot analysis , 2017 .

[8]  J. L. Peterson,et al.  Deep Neural Network Initialization With Decision Trees , 2017, IEEE Transactions on Neural Networks and Learning Systems.

[9]  D. Rubinfeld,et al.  Hedonic housing prices and the demand for clean air , 1978 .

[10]  Nitish Srivastava,et al.  Dropout: a simple way to prevent neural networks from overfitting , 2014, J. Mach. Learn. Res..

[11]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[12]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[13]  Daniel Casey,et al.  The high velocity, high adiabat, ``Bigfoot'' campaign and tests of indirect-drive implosion scaling , 2017 .

[14]  C. Thomas,et al.  BigFoot: a program to reduce risk for indirect drive laser fusion , 2016 .

[15]  S Lebedev,et al.  The Physics of Inertial Fusion , 2004 .

[16]  Yg,et al.  Dropout as a Bayesian Approximation : Insights and Applications , 2015 .

[17]  J. Meyer-ter-Vehn,et al.  The physics of inertial fusion - Hydrodynamics, dense plasma physics, beam-plasma interaction , 2004 .

[18]  Zoubin Ghahramani,et al.  Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning , 2015, ICML.

[19]  Peter Green,et al.  Markov chain Monte Carlo in Practice , 1996 .

[20]  C. J. Cerjan,et al.  Analysis of the neutron time-of-flight spectra from inertial confinement fusion experiments , 2015 .

[21]  Ryan Nora,et al.  Using ensembles of simulations to find high-fidelity post-shot models of inertial confinement implosions at the National Ignition Facility , 2016 .

[22]  Ronald L. Iman Latin Hypercube Sampling , 2006 .

[23]  D. K. Bradley,et al.  The impact of capsule ``tent'' thickness on interpreting low mode shape , 2013 .

[24]  David H. Munro,et al.  Interpreting inertial fusion neutron spectra , 2016 .

[25]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[26]  H. Chipman,et al.  BART: Bayesian Additive Regression Trees , 2008, 0806.3286.

[27]  Steven W. Haan,et al.  Three-dimensional HYDRA simulations of National Ignition Facility targets , 2001 .

[28]  Mohamed Zaki,et al.  Uncertainty in Neural Networks: Bayesian Ensembling , 2018, ArXiv.

[29]  Jorge Nocedal,et al.  On the limited memory BFGS method for large scale optimization , 1989, Math. Program..