Paper Accepted!

Sep 25, 2024: Our paper titled Measuring Déjà Vu Memorization Efficiently has been accepted at NeurIPS 2024.

Paper Accepted!

Sep 25, 2024: Our paper titled Déjà Vu Memorization in Vision–Language Models has been accepted at NeurIPS 2024.

Paper Accepted!

Jul 10, 2023: Our paper titled Combing for Credentials: Active Pattern Extraction from Smart Reply has been accepted at IEEE S&P 2024.

Academic Milestone!

Dec 02, 2022: I passed my PhD Dissertation Defense and am officially a Doctor of Philosophy (Computer Science).

Paper Accepted!

Aug 26, 2022: Our paper titled Are Attribute Inference Attacks Just Imputation? has been accepted at CCS 2022.

Paper Accepted!

Dec 13, 2020: Our paper titled Revisiting Membership Inference Under Realistic Assumptions has been accepted at PETS 2021.

Paper Accepted!

Jun 1, 2019: Our paper titled Evaluating Differentially Private Machine Learning in Practice has been accepted at USENIX Security 2019, Santa Clara, CA, USA.

Paper Accepted!

Sep 5, 2018: Our paper titled Distributed Learning without Distress: Privacy-Preserving Empirical Risk Minimization has been accepted at NeurIPS 2018, Montreal, Canada.

Academic Milestone!

Feb 15, 2018: I finished my PhD Qualifying Examination and now I am a full time PhD Student.

Previous Next

Skills

Python

Statistics

Java

Github

Bitbucket

AWS

Experience

 
 
 
 
 

Postdoctoral Researcher

Meta FAIR

Jan 2023 – Present Menlo Park
Research on privacy preserving machine learning.
 
 
 
 
 

Research Intern

Microsoft Redmond Lab

May 2021 – Aug 2021 Seattle
Research on extracting data from generative language models.
 
 
 
 
 

Graduate Research Assistant

University of Virginia

Aug 2016 – Dec 2022 Virginia
Research on privacy preserving machine learning.
 
 
 
 
 

Research and Development Senior Analyst

Accenture Technology Labs

Jan 2015 – Jul 2016 Bangalore
Application of machine learning for solving software engineering problems.

Awards and Achievements

Awarded Student Travel Grant for presenting at USENIX Security 2019

USENIX Security Aug 2019

Awarded Student Travel Grant for presenting at NeurIPS 2018

NeurIPS Dec 2018

Filed 3 Patents in Accenture Technology Labs

Accenture Technology Labs Jan 2015 – Jul 2016

Recent Posts

Models leak distribution information that can be used for attribute inference.
5 min read

Results for the ongoing work on attribute inference defense.
Last updated on 14 min read

Membership inference attacks are effective even for skewed priors.
5 min read

A hypothesis testing approach to evaluate the Xbox One X HDD.
4 min read

What seems safe, might not be safe in practice.
Last updated on 4 min read

Projects

*
All Machine Learning Differential Privacy Multi-Party Computation

Inference Privacy Risks of Machine Learning Models

Comparing the differential privacy implementations by quantifying their privacy leakage.

Decentralized Certificate Authorities

Allowing certificate authorities to sign digital certificates in a secure and distributed way.

Privacy Preserving Machine Learning

Combining differential privacy and multi-party computation techniques for private machine learning.

Recent Publications

Quickly discover relevant content by filtering publications.

We propose a white-box attribute inference attack that is able to extract distribution information from the model.
PDF Project

We propose a white-box attribute inference attack that is able to extract distribution information from the model.
PDF Code Project Slides Video

We propose black-box and gray-box active pattern extraction attacks that extract sensitive data patters from the Smart Reply model.
PDF Project

We propose novel membership inference attack and a threshold selection procedure to improve the existing attacks.
PDF Code Project Slides

We propose differentially private algorithm for non-convex empirical risk minimization with reduced gradient complexity.
PDF Project
See all publications

Pen Sketches