Adversarial Numerical Analysis for Inverse Problems

10/15/2019
by   Kailai Xu, et al.
0

Many scientific and engineering applications are formulated as inverse problems associated with stochastic models. In such cases the unknown quantities are distributions. The applicability of traditional methods is limited because of their demanding assumptions or prohibitive computational consumptions; for example, maximum likelihood methods require closed-form density functions, and Markov Chain Monte Carlo needs a large number of simulations. We introduce adversarial numerical analysis, which estimates the unknown distributions by minimizing the discrepancy of statistical properties between observed random process and simulated random process. The discrepancy metric is computed with a discriminative neural network. We demonstrated numerically that the proposed methods can estimate the underlying parameters and learn complicated unknown distributions.

READ FULL TEXT

page 20

page 21

page 22

research
12/03/2021

Generalized Transitional Markov Chain Monte Carlo Sampling Technique for Bayesian Inversion

In the context of Bayesian inversion for scientific and engineering mode...
research
11/05/2021

Sampling Methods for Bayesian Inference Involving Convergent Noisy Approximations of Forward Maps

We present Bayesian techniques for solving inverse problems which involv...
research
05/08/2021

Spline-Based Bayesian Emulators for Large Scale Spatial Inverse Problems

A Bayesian approach to nonlinear inverse problems is considered where th...
research
03/10/2021

A Variational Inference Framework for Inverse Problems

We present a framework for fitting inverse problem models via variationa...
research
11/02/2017

Consistent estimation of the spectrum of trace class data augmentation algorithms

Markov chain Monte Carlo is widely used in a variety of scientific appli...

Please sign up or login with your details

Forgot password? Click here to reset