About

Hi! I’m Xiyang Liu, a privacy engineer at Snap Inc.. I work on in privacy by design for generative AI and applied machine learning. I earned my Ph.D. in Computer Science from the University of Washington, advised by Sewoong Oh. My research has focused on the foundations of machine learning, including topics such as differential privacy, secure and robust machine learning.

Feel free to reach out if you’d like to discuss research, collaboration, or any other opportunities.

📚 Publications

Insufficient Statistics Perturbation: Stable Estimators for Private Least Squares
Gavin Brown, Jonathan Hayase, Samuel Hopkins, Weihao Kong, Xiyang Liu, Sewoong Oh, Juan C. Perdomo, Adam Smith
COLT 2024

Label Robust and Differentially Private Linear Regression: Computational and Statistical Efficiency
Xiyang Liu, Prateek Jain, Weihao Kong, Sewoong Oh, Arun Sai Suggala
NeurIPS 2023

DP-PCA: Statistically Optimal and Differentially Private PCA
Xiyang Liu, Weihao Kong, Prateek Jain, Sewoong Oh
NeurIPS 2022

Differential Privacy and Robust Statistics in High Dimensions
Xiyang Liu, Weihao Kong, Sewoong Oh
COLT 2022; [Spotlight talk at PPAI’22 workshop]; [Talk by Sewoong]

Reconstruction of visual images from mouse retinal ganglion cell spiking activity using convolutional neural networks
Tyler Benster, Darwin Babino, John Thickstun, Matthew Hunt, Xiyang Liu, Zaid Harchaoui, Sewoong Oh, Russell N. Van Gelder
bioRxiv

Mace: A flexible framework for membership privacy estimation in generative models
Yixi Xu, Sumit Mukherjee, Xiyang Liu, Shruti Tople, Rahul M Dodhia, Juan M Lavista Ferres
Transactions on Machine Learning Research (TMLR)

Robust and Differentially Private Mean Estimation
Xiyang Liu, Weihao Kong, Sham Kakade, Sewoong Oh
NeurIPS 2021; [Code]; [Talk by Sewoong]; [Talk at FL-ICML’21 (from 45 minutes)]

KO codes: Inventing Nonlinear Encoding and Decoding for Reliable Wireless Communication via Deep-Learning
Ashok Vardhan Makkuva*, Xiyang Liu*, Mohammad Vahid Jamali, Hessam Mahdavifar, Sewoong Oh, Pramod Viswanath
ICML 2021; [Code]; [Talk by Hessam]

Reed-Muller Subcodes: Machine Learning-Aided Design of Efficient Soft Recursive Decoding
Mohammad Vahid Jamali, Xiyang Liu, Ashok Vardhan Makkuva, Hessam Mahdavifar, Sewoong Oh, Pramod Viswanath
ISIT 2021

Minimax Rates of Estimating Approximate Differential Privacy
Xiyang Liu, Sewoong Oh
NeurIPS 2019; [Code]; [Code in PyDP library by OpenMined]

🎓 Dissertation

Privacy meets Robustness: Unveiling the interplay between Differential Privacy and Robustness in Machine Learning
Ph.D. Dissertation, University of Washington, 2024