B.S. in University of Waterloo, Ontario, Canada, 2020–2025
University of Waterloo, Mechanical and Mechatronics Engineering Department
Developed a novel GNN-based method for river ice classification. Presented at Geoinformatics 2024, gave a 20-minute talk, and volunteered at the conference.
Supervisor: Professor Andrea Scott
University of Waterloo, Applied Math Department
Integrated first-order PDEs into GNNs to enhance performance and mitigate over-smoothing.
Supervisor: Professor Hans De Sterck
Aqrose Technology
Developed a new algorithm for detecting data matrix codes. Gained experience in AI and machine learning for computer vision.
Chinese Academy of Sciences, Institute of Semiconductors
Conducted underwater image processing research and implemented a novel dehazing method in C++ using OpenCV.
Learnable Weight Graph Neural Network for River Ice Classification
Qu, Y.; Soleymani, A.; Sudom, D.; Scott, K.A. Proceedings 2024, 110, 30. Link
Enhanced Underwater Image Dehazing Using an Improved Dark Channel Prior Method. Link
First-order PDEs for Graph Neural Networks: Advection and Burgers Equation Models. Link
← Back to Home