Matthew Joseph

matgtjoseph@gmail.com

I'm a research scientist at Google New York, mostly working on differential privacy. Before that, I did a PhD in computer science at UPenn, advised by the excellent Aaron Roth. Before that, I did my undergrad in math at UChicago.

Some Recent Papers

All papers are alphabetical author order.
For a complete list, check Google Scholar.

• Private, Efficient, and Optimal K-Norm and Elliptic Gaussian Noise For Sum, Count, and Vote
  With Alexander Yu
  Manuscript.
  
  
• Better Private Linear Regression Through Better Private Feature Selection
  With Travis Dick and Jennifer Gillenwater
  NeurIPS 2023
  
  
• Easy Differentially Private Linear Regression
  With Kareem Amin, Mónica Ribero, and Sergei Vassilvitskii
  ICLR 2023
  
  

Service

• PC: FAT* 2019; TPDP 2019, 2020, 2022.
• Conference Reviews: AIES 2019; COLT 2017; EC 2018; FOCS 2019; ICLR 2020, 2022, 2023; ICML 2019, 2020 (top reviewer!), 2022, 2023; ITC 2020; ITCS 2021; NeurIPS 2016, 2018 (top reviewer!), 2019, 2020, 2021, 2022, 2023 (and ethics reviewer); SODA 2021, 2022; STOC 2020.
• Journal Reviews: Journal of Machine Learning Research 2020; Journal of Privacy and Confidentiality 2019, 2021.

Other

I also enjoy cooking, falling off things, Wikipedia's random page function, and this website.