Yu Wei

PhD student in Computer Science

prof_pic.jpg

Georgia Tech

756 West Peachtree Street NW, Atlanta, GA

My name is Wei, Yu (卫煜). I am a PhD student in the College of Computing at the Georgia Institute of Technology, studying cryptography under the supervision of Professor Vassilis Zikas. Before transferring to Georgia Tech, I pursued my PhD at Purdue University. Prior to that, I earned a Master’s degree in Computer Science, a Bachelor’s degree in Engineering, and a Bachelor’s degree in Law from Nankai University in China. Regarding work, my passion is to help people benefit from vast potential of data while protect the privacy of those who contribute it. I am particularly interested in exploring the intersection of machine learning, cryptography, and privacy.

My recent research focuses on developing generic tools for differential privacy analysis, with a central question: How to compute the privacy profile of an arbitrary randomized algorithm that meets minimal assumptions? I approach this question through various lens, aiming to build an easy-to-use privacy analysis toolkit for domain experts – those who know well how to extract valuable insights from data but may not be familiar with the intricacies of proving privacy guarantees. Beyond this, there are a lots of particular theoretical/pratical questions I want to explore, for details please feel free to reach out and discuss with me!

I am also fortunate to work with researchers in various fields, including secure computation, machine learning, symmetric key cryptography, and game theory. It is a real pleasure for me to learn from and exchange ideas at any intersection of privacy and various disciplines. If you are interested in working with me, please do not hesitate to reach out. I am eager to exchange ideas and welcome your input :smile: .

news

Aug 31, 2024 I was excited to have our paper on information-theoretic mluti-server PIR accepted at TCC 2024!
Jul 29, 2024 I spent this summer as a visiting researcher at UMD, and I want to give a huge thanks to Professor Katz for being an amazing host! Next up, I’ll be heading to Georgia Tech to continue my PhD journey with Professor Zikas. I’m super grateful for everything that’s happened so far and can’t wait to see what the future holds!
Mar 19, 2024 I was excited to have my paper on black-box differential privacy estimators accepted at S&P 2024!
Mar 05, 2024 Please check my work "The Normal Distributions Indistinguishability Spectrum and its Application to Privacy-Preserving Machine Learning" on arXiv.
Oct 09, 2023 I was excited to win the 2023 Korea National Cryptography Contest First Prize!

selected publications

  1. S&P
    Eureka: A General Framework for Black-box Differential Privacy Estimators
    Yun Lu , Malik Magdon-Ismail , Yu Wei, and Vassilis Zikas
    In 2024 IEEE Symposium on Security and Privacy (SP) , May 2024
  2. Distributed Differential Privacy via Shuffling Versus Aggregation: A Curious Study
    Yu Wei, Jingyu Jia , Yuduo Wu , Changhui Hu , Changyu Dong , and 4 more authors
    Trans. Info. For. Sec., Jan 2024
  3. How to Make Private Distributed Cardinality Estimation Practical, and Get Differential Privacy for Free
    Changhui Hu , Jin Li , Zheli Liu , Xiaojie Guo , Yu Wei, and 3 more authors
    In USENIX Security Symposium , Jan 2021