About me

I am a passionate and enthusiastic person, I work well both as a team and on my own. I love to engage technical audiences to influence them with my research. I express empathy to Microsoft’s slogan, “to empower every person and every organization on the planet to achieve more”, since I belive the essence of engineering is to enfluence everyday lives of various people.

I am currently a Master’s student at the Graduate School of Information Security. As a researcher in WSP Lab, my research mainly focuses on safe AI.

Research Experiences

Web Security & Privacy Laboratory, KAIST | advisor: Prof. Sooel Son
Master’s Student (2021.09 - present)

  • currently focusing on Safe-AI

Information System Security Laboratory, Korea University | advisor: Prof. Junbeom Hur
Undergraduate Research Intern (2019.03 – 2021.08)

  • Analyzed microarchitectural vulnerabilities of the CPU
  • Developed neural network stealing algorithm leveraging microarchitectural vulnerabilities

Artificial Intelligence Laboratory, Korea University | advisor: Prof. Dongsuk Yook
Undergraduate Research Intern (2018.12 – 2019.02)

  • Implemented model-parallel systems for extremely large deep neural networks

Projects & Accomplishments

Undergraduate Research Program (12th Creative Challenger Program) (2020.04 – 2021.02)

  • Description: 부채널 공격으로 인한 클라우드 취약점 분석
  • Contribution(s): Investigation of potential public Cloud vulnerabilities caused by side-channel attacks, developed an algorithm to steal neural networks from deep learning services running in the cloud by leveraging microarchitectural vulnerabilities
  • Output: video, paper, Cyber Security Paper Competition Grand Prize

Lab Intern Project (2019.09 – 2019.12)

  • Desciption: Disclosing Neural Network Structure for Better Adversarial Attacks
  • Contribution(s): Neural network structure inference from side-channels to construct stronger adversarial attacks
  • Output: Final Report

K-Startup Challenge (2019.06 – 2019.08)

  • Description: Stock Plantation: Robo-advisor Crowdfunding Platform
  • Contribution(s): Idea proposal, implementation of RL & LSTM-based AI for automated stock investment, applied NLP techniques to use news articles keywords for stock price prediction
  • Output: server-side code, robo-advisor code, Promising Start-up Team Award

Undergraduate Research Program (11th Creative Challenger Program) (2019.04 – 2020.02)

  • Description: 다양한 관점의 뉴스 클러스터에서 객관적 정보 분리 및 논조 파악
  • Contribution(s): Algorithm development for political stance classification of Internet news articles
  • Output: code, video

Intel AI Drone Hackathon (2018.11.08 – 2018.11.09)

  • Description: Adaptive public WiFi using drones as WAPs
  • Contribution(s): Idea proposal, implementation of object detection and traffic clustering
  • Output: Best Innovation Award