Jinshuo Dong

alt text 

About me:
I am currently a post-doc at IDEAL (i.e. Northwestern University and TTI Chicago). Previously, I got my PhD in applied math (AMCS) from the University of Pennsylvania, under the supervision of Aaron Roth. Prior to Penn, I obtained my bachelor's degree in mathematics from Peking University in 2014. My CV can be found here.

Email: first name@northwestern.edu


Current focus of my work is differential privacy and its role in statistics and machine learning. I'm also generally interested in expanding my knowledge in the mathematics of information and computation.


  • A Central Limit Theorem for Differentially Private Query Answering
    Jinshuo Dong, Weijie Su and Linjun Zhang.
    To appear in NeurIPS 2021. [arXiv]

  • Optimal Accounting of Differential Privacy via Characteristic Function
    Yuqing Zhu, Jinshuo Dong and Yu-Xiang Wang.
    In submission. [arXiv]

  • Privacy Amplification via Iteration for Shuffled and Online PNSGD
    Matteo Sordello, Zhiqi Bu, Jinshuo Dong.
    In the proceedings of ECML/PKDD 2021.
    [arXiv] [proceeding]

  • Rejoinder: Gaussian Differential Privacy
    Jinshuo Dong, Aaron Roth and Weijie Su.
    Discussion reply. J R Stat Soc Series B. 2021;00:1–4. [arXiv]

  • Optimal Differential Privacy Composition for Exponential Mechanisms
    Jinshuo Dong, David Durfee and Ryan Rogers.
    In the proceedings of ICML 2020.
    [blog post] [arXiv] [proceeding]

  • Sharp Composition Bounds for Gaussian Differential Privacy via Edgeworth Expansion
    Qinqing Zheng, Jinshuo Dong, Qi Long, and Weijie Su.
    In the proceedings of ICML 2020.
    [arXiv] [proceeding]

  • Equilibrium Characterization for Data Acquisition Games.
    Jinshuo Dong, Hadi Elzayn, Shahin Jabbari, Michael Kearns and Zachary Schutzman.
    In the proceedings of IJCAI 2019.
    [arXiv] [proceeding]

  • Strategic Classification from Revealed Preferences
    Jinshuo Dong, Aaron Roth, Zachary Schutzman, Bo Waggoner and Steven Wu.
    In the proceedings of EC 2018.
    [arXiv] [proceeding]