Wei Guo

Wei Guo

Welcome to my homepage! I received my PhD in industrial and systems engineering from University of Washington in 2020. During my PhD years, I mostly worked as a research assistant for Boeing Advanced Research Center, focusing on exploring the use of topological data analysis in computer vision and dynamic networks. During that time, I have partaken in multiple Boeing projects aimed at revamping manufacturing processes with state‐of‐the‐art machine learning solutions for increased efficiency. One project where I played a major role, titled “ Deep learning for automated in-process inspection of composite layup,” was recognized by winning the best presentation award in data analytics track at the Boeing Tech Excellence Conference in 2019. A US patent application is also being filed for this work.

In addition, I am passionate about the application of deep learning to autonomous systems. The field of autonomous systems also reconnects the experience I have gained in resolving trajectory optimization problems of aerial vehicles in winds / obstacle fields during my graduate studies at University of Minnesota. I am currently in a transition from academia to industry. I expect to start working from May, 2020.

Interests

  • Computer Vision
  • Deep Learning
  • Data Analytics
  • Autonomous Systems

Education

  • PhD in Industrial and Systems Engineering, 2020

    University of Washington

  • MSc in Industrial and Systems Engineering, 2014

    University of Minnesota

  • MSc in Aerospace Engineering and Mechanics, 2010

    University of Minnesota

  • MSc in Control Science and Engineering, 2008

    Harbin Institute of Technology

  • BSc in Control Science and Engineering, 2006

    Harbin Institute of Technology

Featured Publications

Wei Guo, Krithika Manohar, Steven L. Brunton and Ashis G. Banerjee
IEEE Transactions on Knowledge and Data Engineering, 30(7): 1403-1408

Sparse-TDA: Sparse Realization of Topological Data Analysis for Multi-Way Classification

We present a new method, referred as Sparse-TDA method, that combines persistence image-based TDA method with QR pivoting-based sparse sampling algorithm to transform topological features into image pixels and identify discriminative pixel samples in the presence of noisy and redundant information.

[+/-] Key figure
pdf code
Wei Guo and Ashis G. Banerjee
Journal of Manufacturing Systems, 43(2): 225-234

Identification of Key Features Using Topological Data Analysis for Accurate Prediction of Manufacturing System Outputs

We apply TDA for the first time (to the best of our knowledge) on manufacturing process data to extract the key process variables that have the most significant impact on the outputs. The results are summarized in the form of a topological network to facilitate a better understanding of the casual relationships between process variables and outputs through direct visualization.

[+/-] Key figure
pdf

Working Paper

Wei Guo, Ruqian Chen, Yen-Chi Chen and Ashis G. Banerjee
Working paper

Efficient Community Detection in Large-Scale Dynamic Networks Using Topological Data Analysis

We develop a framework for analyzing communitity structures in dynamic networks based on the established community tree representation. With the information revealed by community trees and the corresponding persistence diagrams, our proposed approach can efficiently detect clique communities and keep track of the major structural changes during their evolution given a stability threshold.

[+/-] Key figure
code