An efficient Bayesian neural network surrogate algorithm for shape detection

M. Ganesh, S. C. Hawkins, N. Kordzakhia, S. Unicomb

Research output: Contribution to journalConference paperpeer-review


We present an efficient Bayesian algorithm for identifying the shape of an object from noisy far field data. The data is obtained by illuminating the object with one or more incident waves. Bayes' theorem provides a framework to find a posterior distribution of the parameters that determine the shape of the scatterer. We compute the distribution using the Markov Chain Monte Carlo (MCMC) method with a Gibbs sampler. The principal novelty of this work is to replace the forward far-field-ansatz wave model (in an unbounded region) in the MCMC sampling with a neural-network-based surrogate that is hundreds of times faster to evaluate. We demonstrate the accuracy and efficiency of our algorithm by constructing the distributions, medians and confidence intervals of non-convex shapes using a Gaussian random circle prior.
Original languageEnglish
Pages (from-to)C112-C127
Number of pages16
JournalThe ANZIAM Journal
Publication statusPublished - 2 Sep 2020
EventComputational Techniques and Applications Conference (20th : 2020) - Online
Duration: 30 Aug 20202 Sep 2020


Dive into the research topics of 'An efficient Bayesian neural network surrogate algorithm for shape detection'. Together they form a unique fingerprint.

Cite this