Soo Min Kwon
 |
Soo Min Kwon Curriculum Vitae (Last Updated 10/07/2024) |
Recent News
A paper, BLAST: Block-Level Adaptive Structured Matrices for Efficient Deep Neural Network Inference, is accepted to NeurIPS 2024!
A paper, Efficient Compression of Overparameterized Deep Models through Low-Dimensional Learning Dynamics, is accepted to AISTATS 2024!
About Me
I am a PhD candidate in the EECS Department at the University of Michigan-Ann Arbor, co-advised by Prof. Laura Balzano and Prof. Qing Qu. I completed my M.S. degree in the ECE Department at Rutgers University under Prof. Anand D. Sarwate, researching in learning from low-rank approximations of tensor data. I also obtained my B.S. degree in the ECE Department at Rutgers University with high honors and a minor in Mathematics.
Lately, I am interested in generative models (e.g. diffusion models) as well as overparameterized models (e.g. deep neural networks) and the inherent bias in their learning dynamics towards low-rank solutions.
I have recently picked up running as a hobby.
Education
Ph.D. in Electrical and Computer Engineering, University of Michigan-Ann Arbor, Sept. 2022 - Present
M.S. in Electrical and Computer Engineering, Rutgers University-New Brunswick, Sept. 2020 - May 2022
B.S. in Electrical and Computer Engineering, Rutgers University-New Brunswick, Sept. 2016 - May 2020