MANIFOLD LEARNING-BASED POLYNOMIAL CHAOS EXPANSIONS FOR HIGH-DIMENSIONAL SURROGATE MODELS

Katiana Kontolati, Dimitrios Loukrezis, Ketson R.M. Dos Santos, Dimitrios G. Giovanis, Michael D. Shields

Research output: Contribution to journalArticlepeer-review

17 Scopus citations

Abstract

In this work we introduce a manifold learning-based method for uncertainty quantification (UQ) in systems describing complex spatiotemporal processes. Our first objective is to identify the embedding of a set of high-dimensional data representing quantities of interest of the computational or analytical model. For this purpose, we employ Grassmannian diffusion maps, a two-step nonlinear dimension reduction technique which allows us to reduce the dimensionality of the data and identify meaningful geometric descriptions in a parsimonious and inexpensive manner. Polynomial chaos expansion is then used to construct a mapping between the stochastic input parameters and the diffusion coordinates of the reduced space. An adaptive clustering technique is proposed to identify an optimal number of clusters of points in the latent space. The similarity of points allows us to construct a number of geometric harmonic emulators which are finally utilized as a set of inexpensive pretrained models to perform an inverse map of realizations of latent features to the ambient space and thus perform accurate out-of-sample predictions. Thus, the proposed method acts as an encoder-decoder system which is able to automatically handle very high-dimensional data while simultaneously operating successfully in the small-data regime. The method is demonstrated on two benchmark problems and on a system of advection-diffusion-reaction equations which model a first-order chemical reaction between two species. In all test cases, the proposed method is able to achieve highly accurate approximations which ultimately lead to the significant acceleration of UQ tasks.

Original languageEnglish (US)
Pages (from-to)39-64
Number of pages26
JournalInternational Journal for Uncertainty Quantification
Volume12
Issue number4
DOIs
StatePublished - 2022
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2022 by Begell House, Inc. www.begellhouse.com.

Keywords

  • advection-diffusion-reaction
  • Grassmann manifold
  • large-scale computational systems
  • low-dimensional embedding
  • manifold learning
  • surrogate modeling
  • uncertainty quantification

Fingerprint

Dive into the research topics of 'MANIFOLD LEARNING-BASED POLYNOMIAL CHAOS EXPANSIONS FOR HIGH-DIMENSIONAL SURROGATE MODELS'. Together they form a unique fingerprint.

Cite this