Changwoo Lee
Machine Learning Researcher at Qualcomm
Ann Arbor, Michigan, United States
Summary
👤
Senior🎓
Top SchoolChangwoo Lee is a PhD candidate and Graduate Student Research Assistant at the University of Michigan with nine years of experience focused on making deep learning models smaller, faster, and more robust. His research specializes in DNN compression, learnable structural sparsity, and efficient multimodal systems, with internships at Samsung Research America and Google DeepMind that bridge academic rigor and industry-scale challenges. He combines electrical and computer engineering foundations from Hanyang University with practical ML research, often exploring methods that reduce deployment cost without sacrificing accuracy. Known for pursuing efficiency at the algorithm-architecture boundary, he aims to enable resource-constrained and real-world multimodal applications.
10 years of coding experience
6 years of employment as a software developer
Doctor of Philosophy - PhD, Electrical and Computer Engineering, Doctor of Philosophy - PhD, Electrical and Computer Engineering at University of MichiganMaster of Science - MS, Electronics and Computer Engineering, Master of Science - MS, Electronics and Computer Engineering at Hanyang University
English, Korean
Github Skills (13)
vivado10
accelerator10
fpga-accelerator10
vivado-hls9
generative-adversarial-network7
python5
error-correction5
tensorflow4
deep-learning3
pytorch3
jekyll-theme3
deep-neural-networks3
jekyll1
Programming languages (3)
Github contributions (5)
changwoolee/lenet5_hls
May 2017 - May 2018
FPGA Accelerator for CNN using Vivado HLS
Contributions:1 release, 62 commits, 2 PRs in 1 year
fpga-acceleratorvivado-hlsvivadolenethigh-level-synthesis
changwoolee/Gaudi-GBLR
Jan 2024 - Sep 2024
An official PyTorch implementation of "Differentiable Learning of Generalized Structured Matrices for Efficient Deep Neural Networks," ICLR 2024.
Contributions:1 release, 5 pushes, 1 tag in 7 months
deep-neural-networkspytorch
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.