Shanshan Wu

As of Sept 2019, I got my PhD from the ECE Department at the University of Texas at Austin. I am fortunate to be advised by Sujay Sanghavi and Alex Dimakis. My next stop is Google Research, with a focus on Federated Learning. Prior to PhD, I graduated from Shanghai Jiao Tong University, advised by Xudong Wang.

My PhD thesis is Unsupervised Learning for Large-Scale Data. Code is available at my GitHub. I can be reached at wushanshan0701@gmail.com.

Last update: April 9, 2024

Publications

Prompt Public Large Language Models to Synthesize Data for Private On-device Applications
Shanshan Wu*, Zheng Xu*, Yanxiang Zhang*, Yuanbo Zhang, Daniel Ramage
Conference on Language Modeling (COLM) 2024
*Equal contribution

Profit: Benchmarking Personalization and Robustness Trade-off in Federated Prompt Tuning
Liam Collins, Shanshan Wu, Sewoong Oh, Khe Chai Sim
Short version in NeurIPS Workshop on Federated Learning, 2023 (Best Paper)

Motley: Benchmarking Heterogeneity and Personalization in Federated Learning
[Code]
Shanshan Wu, Tian Li, Zachary Charles, Yu Xiao, Ziyu Liu, Zheng Xu, Virginia Smith
Short version in NeurIPS Workshop on Federated Learning, 2022

A Field Guide to Federated Optimization
[Code]
Joint effort of 50+ authors: Jianyu Wang, …, Shanshan Wu, …, 2021

Federated Reconstruction: Partially Local Federated Learning
[Tutorial]
Karan Singhal, Hakim Sidahmed, Zachary Garrett, Shanshan Wu, Keith Rush, Sushant Prakash
Advances in Neural Information Processing Systems (NeurIPS) 2021

Implicit Regularization and Convergence for Weight Normalization
[Slides]
Xiaoxia Wu*, Edgar Dobriban*, Tongzheng Ren*, Shanshan Wu*, Yuanzhi Li, Suriya Gunasekar, Rachel Ward, Qiang Liu
Advances in Neural Information Processing Systems (NeurIPS) 2020
*Equal contribution

Learning Distributions Generated by One-Layer ReLU Networks
[Code] [Poster] [Slides]
Shanshan Wu, Alex Dimakis, Sujay Sanghavi
Advances in Neural Information Processing Systems (NeurIPS) 2019

Sparse Logistic Regression Learns All Discrete Pairwise Graphical Models
[Code] [Poster] [Slides]
Shanshan Wu, Sujay Sanghavi, Alex Dimakis
Advances in Neural Information Processing Systems (NeurIPS) 2019 (Spotlight)
Short version in NeurIPS Workshop on Relational Representation Learning, 2018

Learning a Compressed Sensing Measurement Matrix via Gradient Unrolling
[Code] [Poster] [Slides]
Shanshan Wu, Alex Dimakis, Sujay Sanghavi, Felix X. Yu, Dan Holtmann-Rice, Dmitry Storcheus, Afshin Rostamizadeh, and Sanjiv Kumar
International Conference on Machine Learning (ICML) 2019

Single Pass PCA of Matrix Products
[Code] [Poster] [Spotlight Videl]
Shanshan Wu, Srinadh Bhojanapalli, Sujay Sanghavi, and Alex Dimakis
Advances in Neural Information Processing Systems (NeurIPS) 2016

Leveraging Sparsity for Efficient Submodular Data Summarization
[Spotlight Video]
Erik Lindgren, Shanshan Wu, and Alex Dimakis
Advances in Neural Information Processing Systems (NeurIPS) 2016
Short version in NeurIPS workshop OPT 2015

Distributed Opportunistic Scheduling with QoS Constraints for Wireless Networks with Hybrid Links
Wenguang Mao, Xudong Wang, and Shanshan Wu
IEEE Transactions on Vehicular Technology (TVT), 2015
An earlier version appears in Proceedings of the IEEE Globecom, 2013

Performance Study on a CSMA/CA-Based MAC Protocol for Multi-User MIMO Wireless LANs
[Code]
Shanshan Wu, Wenguang Mao, and Xudong Wang
IEEE Transactions on Wireless Communications (TWC), 2014
An earlier version appears in Proceedings of the IEEE Globecom, 2013

Information-theoretic study on routing path selection in two-way relay networks
Shanshan Wu, Wenguang Mao, and Xudong Wang
Proceedings of the IEEE Globecom, 2013

Internships

June 2018 - August 2018
Software Engineer Intern | Google Research, New York City

June 2017 - August 2017
Software Engineer Intern | Google Research, New York City

Jan 2017 - April 2017
Applied Scientist Intern | Amazon AI, East Palo Alto

Professional services

Conference reviewer:

NeurIPS 2016/2017/2018/2019/2021/2022 (top 30% highest-rating reviewer for NeurIPS 2018)
ICML 2018/2019/2020/2023
ICLR 2022/2024
IEEE ISIT 2019

Journal reviewer:

Transactions on Machine Learning Research (TMLR)
Journal on Machine Learning Research (JMLR)
IEEE Trans. on Mobile Computing
IEEE Trans. on Wireless Communications
IEEE Trans. on Vehicular Technology
Ad Hoc Networks

Graduate Courses at UT-Austin

2016 Fall
CS395T Sublinear Algorithms (Prof. Eric Price)
Course project: Rescaled JL Embedding

2016 Spring
EE381K-6 Estimation Theory (Prof. Haris Vikalo)
Course project: A Survey of Fast Kernel Sketching Algorithms

2015 Fall
EE381V Advanced Probability in Learning, Inference, and Networks (Prof. Sanjay Shakkottai)
Course project: Low-Rank Approximation of Matrix Product in One Pass
CS388G Algorithms: Techniques/Theory (Prof. Vijaya Ramachandran)
Course project: PTAS for the Euclidean Traveling Salesman Problem

2015 Summer (Online courses provided by edX)
CS100.1x Introduction to Big Data with Apache Spark (Prof. Anthony D. Joseph) Certificate
CS190.1x Scalable Machine Learning (Prof. Ameet Talwalkar) Certificate

2015 Spring
EE381V Advanced Algorithms (Prof. Evdokia Nikolova)
Course project: Signal Recovery from Permuted Observations
EE381K Information Theory (Prof. Alex Dimakis)

2014 Fall
EE380L Data Mining (Prof. Joydeep Ghosh)
Course project: Ranking by Alternating SVM and Factorization Machine
EE381V Large-Scale Optimization (Prof. Sujay Sanghavi)
EE381J Probability and Stochastic Processes (Prof. Sanjay Shakkottai)

Teaching Experiences

Teaching Assistant, EE381V (Machine Learning for Large Scale Data), UT-Austin, Spring 2016.
Teaching Assistant, EE313 (Linear Systems and Signals), UT-Austin, Fall 2014.
Teaching Assistant, VE489 (Computer Networks), UM-SJTU Joint Institute, Summer 2013.
Teaching Assistant, VP140 (Physics I), UM-SJTU Joint Institute, Summer 2009.