Yiqun Xie

PhD candidate
Computer Science and Engineering
University of Minnesota, Twin Cities
Email: xiexx347 at

Advisor: Dr. Shashi Shekhar

I am currently a PhD candidate in Computer Science at the University of Minnesota, Twin-Cities.

General Research Fields: Optimization, Spatial Computing, Machine Learning, Data Mining

Current Research: Optimization and machine learning in geospatial domain (e.g., space allocation, geospatial object detection)

  • Spatial optimization: Space partitioning and allocation techniques which can be applied in landscape design, harvest scheduling, urban planning, floor zoning, etc. The problems are often combinatorial in nature involving both continuous space partitioning and multiple-choice assignment. My research carries out exact algorithms, approximation algorithms and heuristic algorithms to solve the challenging spatial allocation problems. In addition, I am also exploring data-driven approaches (e.g., machine learning) to efficiently solve large-size real-world problems despite their theoretical hardness (e.g., NP-hard, APX-hard, NPO-complete).
  • Spatial deep learning: Deep networks (e.g., convolutional neural net and its variations) to identify geospatial objects. Investigate the success and failure modes of the frameworks. More coming.

Before joining the spatial computing group, I mainly worked on pattern recognition in remote sensing imagery and LiDAR point cloud. For example, I have published algorithms for crater detection on the lunar surface which potentially extends the currently largest lunar crater catalogue LU60645GT. During my internship at Esri (a spatial software leader), I worked on building footprint and tree canopy recognition using machine learning.

Besides research work, I was also involved in both grad-level and undergrad-level teaching. I have worked as the teaching assistant for both introductory and advanced courses on machine learning, spatial computing and database systems.