You are cordially invited to my thesis defense on March 29th.

Title: New Gaussian Process Modeling for Low-rank and Simulated Data

Date: March 29th, Wednesday

Time: 12.00pm - 1.30pm

Location: ISyE Groseclose 118

Zoom link: https://gatech.zoom.us/j/5109543006

Henry Shaowu Yuchi

Machine Learning PhD Student

Industrial and Systems Engineering
Georgia Institute of Technology

Committee

1 Professor C. F. Jeff Wu (Advisor), Industrial and Systems Engineering, Georgia Tech

2 Professor Yao Xie (Advisor), Industrial and Systems Engineering, Georgia Tech

3 Professor V. Roshan Joseph, Industrial and Systems Engineering, Georgia Tech

4 Professor Simon Mak, Department of Statistical Science, Duke University

5 Professor Mark Davenport, Electrical and Computer Engineering, Georgia Tech

Abstract

Gaussian process and Gaussian distribution are popular tools for modeling and enjoy a wide range of applications across complex scientific and engineering systems. In this defense, we investigate how the Gaussian process is integrated with the multi-fidelity framework and how matrix-variate Gaussian distribution is utilized to avail matrix completion problems. The defense is composed of the contributions made in these two directions, including (i) a new experimental design approach to multi-fidelity finite element simulations, (ii) a new conglomerate multi-fidelity emulator model utilizing the Gaussian process that enables us to tackle multi-dimensional fidelity parameters, (iii) a new Bayesian matrix completion model utilizing matrix-variate Gaussian process that facilitates subspace estimation and uncertainty quantification, and (iv) a new information recovery framework for piezoresponse microscopy data in material science studies via low-rank matrix completion and uncertainty quantification. We review the current works and demonstrate the new contributions using real-world applications.