Hoki Kim
Assistant Professor at Chung-Ang University
Seoul, South Korea
Summary
🤩
Rockstar🎓
Top SchoolHoki Kim is an Assistant Professor at Chung-Ang University with seven years of experience at the intersection of machine learning research and software engineering. Previously a postdoctoral researcher at Seoul National University, he completed an integrated PhD there and focuses on neural network robustness and adversarial attacks. He has hands-on experience implementing and maintaining PyTorch adversarial attack methods—contributing to popular toolsets by adding MultiAttack and DI2FGSM and modernizing implementations like FGSM, PGD, CW, and DeepFool. Based in Seoul, he combines academic rigor with practical ML engineering, routinely debugging and updating attack code to keep it compatible with current frameworks. His work reflects both deep theoretical knowledge and a pragmatic approach to reproducible, open-source research tools. Colleagues note his knack for turning complex adversarial concepts into reliable, reusable code.
8 years of coding experience
1 year of employment as a software developer
Integrated PhD, Integrated PhD at Seoul National UniversityBachelor, Bachelor at 서울대학교 (Seoul National University)
Github Skills (5)
pytorch10
bim10
machine-learning10
adversarial-attacks10
deep-learning10
Programming languages (4)
Github contributions (5)
Harry24k/adversarial-attacks-pytorch
Apr 2019 - Dec 2022
PyTorch implementation of adversarial attacks [torchattacks]
Role in this project:
ML Engineer
ML EngineerContributions:9 releases, 397 commits, 36 PRs in 3 years 8 months
Contributions summary:Hoki implemented and updated various adversarial attacks in PyTorch, including FGSM, BIM, RFGSM, CW, PGD, and DeepFool, with modifications to also include MultiAttack and DI2FGSM. These implementations involve core machine learning concepts for the purpose of studying the robustness of neural networks. The user contributed to the codebase by adding new attack methods and ensuring that attacks functions correctly by fixing bugs and updating them to the current version.
adversarial-attackspytorchdeep-learning
Harry24k/catastrophic-overfitting
Oct 2020 - Jun 2022
Understanding Catastrophic Overfitting in Single-step Adversarial Training [AAAI 2021]
Contributions:18 commits, 16 pushes, 1 branch in 1 year 9 months
adversarial-attacksadversarial-trainingadversarial-defensecatastrophic-overfitting
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.