Variational Bayesian Optimal Experimental Design with Normalizing Flows

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dong, Jiayuan, Jacobsen, Christian, Khalloufi, Mehdi, Akram, Maryam, Liu, Wanjiao, Duraisamy, Karthik, Huan, Xun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912347902181376
author Dong, Jiayuan
Jacobsen, Christian
Khalloufi, Mehdi
Akram, Maryam
Liu, Wanjiao
Duraisamy, Karthik
Huan, Xun
author_facet Dong, Jiayuan
Jacobsen, Christian
Khalloufi, Mehdi
Akram, Maryam
Liu, Wanjiao
Duraisamy, Karthik
Huan, Xun
contents Bayesian optimal experimental design (OED) seeks experiments that maximize the expected information gain (EIG) in model parameters. Directly estimating the EIG using nested Monte Carlo is computationally expensive and requires an explicit likelihood. Variational OED (vOED), in contrast, estimates a lower bound of the EIG without likelihood evaluations by approximating the posterior distributions with variational forms, and then tightens the bound by optimizing its variational parameters. We introduce the use of normalizing flows (NFs) for representing variational distributions in vOED; we call this approach vOED-NFs. Specifically, we adopt NFs with a conditional invertible neural network architecture built from compositions of coupling layers, and enhanced with a summary network for data dimension reduction. We present Monte Carlo estimators to the lower bound along with gradient expressions to enable a gradient-based simultaneous optimization of the variational parameters and the design variables. The vOED-NFs algorithm is then validated in two benchmark problems, and demonstrated on a partial differential equation-governed application of cathodic electrophoretic deposition and an implicit likelihood case with stochastic modeling of aphid population. The findings suggest that a composition of 4--5 coupling layers is able to achieve lower EIG estimation bias, under a fixed budget of forward model runs, compared to previous approaches. The resulting NFs produce approximate posteriors that agree well with the true posteriors, able to capture non-Gaussian and multi-modal features effectively.
format Preprint
id arxiv_https___arxiv_org_abs_2404_13056
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Variational Bayesian Optimal Experimental Design with Normalizing Flows
Dong, Jiayuan
Jacobsen, Christian
Khalloufi, Mehdi
Akram, Maryam
Liu, Wanjiao
Duraisamy, Karthik
Huan, Xun
Machine Learning
Computational Engineering, Finance, and Science
Computation
Methodology
62K05, 94A17, 62C10, 62F15
Bayesian optimal experimental design (OED) seeks experiments that maximize the expected information gain (EIG) in model parameters. Directly estimating the EIG using nested Monte Carlo is computationally expensive and requires an explicit likelihood. Variational OED (vOED), in contrast, estimates a lower bound of the EIG without likelihood evaluations by approximating the posterior distributions with variational forms, and then tightens the bound by optimizing its variational parameters. We introduce the use of normalizing flows (NFs) for representing variational distributions in vOED; we call this approach vOED-NFs. Specifically, we adopt NFs with a conditional invertible neural network architecture built from compositions of coupling layers, and enhanced with a summary network for data dimension reduction. We present Monte Carlo estimators to the lower bound along with gradient expressions to enable a gradient-based simultaneous optimization of the variational parameters and the design variables. The vOED-NFs algorithm is then validated in two benchmark problems, and demonstrated on a partial differential equation-governed application of cathodic electrophoretic deposition and an implicit likelihood case with stochastic modeling of aphid population. The findings suggest that a composition of 4--5 coupling layers is able to achieve lower EIG estimation bias, under a fixed budget of forward model runs, compared to previous approaches. The resulting NFs produce approximate posteriors that agree well with the true posteriors, able to capture non-Gaussian and multi-modal features effectively.
title Variational Bayesian Optimal Experimental Design with Normalizing Flows
topic Machine Learning
Computational Engineering, Finance, and Science
Computation
Methodology
62K05, 94A17, 62C10, 62F15
url https://arxiv.org/abs/2404.13056