Anshuman Suri

Anshuman Suri

PhD Student

University of Virginia

Biography

I am a third year PhD student in the Computer Science Department at the University of Virginia, advised by Professor David Evans. My primary area of focus is security/privacy in machine learning.

Interests
  • Machine Learning Privacy
  • Adversarial Machine Learning
  • Information Retrieval
Education
  • PhD in Computer Science, 2019-Present

    University of Virginia

  • B.Tech (Hons) in Computer Science, 2014-2018

    IIIT-Delhi

Experience

 
 
 
 
 
University of Virginia
Graduate Research Assistant
Aug 2019 – Present Virginia, USA
Research on privacy and security in Machine Learning at SRG.
 
 
 
 
 
Oracle
Research Assistant (Internship)
Sep 2021 – Present Massachusetts, USA (remote)
Investigating aspects of Privacy in Machine Learning systems and building Federate Learning systems.
 
 
 
 
 
Microsoft
Applied Scientist
May 2018 – Jul 2019 Hyderabad, India
 
 
 
 
 
Microsoft
Software Development Intern
May 2017 – Jul 2017 Hyderabad, India
Worked on making AI-powered game bots, part of Bing STCI.

Recent Posts

Reassessing adversarial training with fixed data augmentation
A recent bug discovery on Pytorch+Numpy got me thinking: how much does this bug impact adversarial robustness?
Reassessing adversarial training with fixed data augmentation

Recent Publications

Quickly discover relevant content by filtering publications.
(2021). Formalizing and Estimating Distribution Inference Risks. In arXiv.

PDF Code

(2021). Formalizing Distribution Inference Risks. In TPDP, ICML 2021.

PDF Cite Code Poster

(2021). Model-Targeted Poisoning Attacks with Provable Convergence. In ICML, 2021.

PDF Cite Code

Contact