• Publications
  • Influence
Membership Inference Attacks Against Machine Learning Models
TLDR
We quantitatively investigate how machine learning models leak information about the individual data records on which they were trained. Expand
  • 1,039
  • 161
  • PDF
Privacy-preserving deep learning
TLDR
In this paper, we present a practical system that enables multiple parties to jointly learn an accurate neural-network model for a given objective without sharing their input datasets. Expand
  • 934
  • 117
  • PDF
De-anonymizing Social Networks
TLDR
We present a framework for analyzing privacy and anonymity in social networks and develop a new re-identification algorithm targeting anonymized social-network graphs. Expand
  • 1,221
  • 113
  • PDF
Robust De-anonymization of Large Sparse Datasets
TLDR
We present a new class of statistical de- anonymization attacks against high-dimensional micro-data, such as individual preferences, recommendations, transaction records and so on. Expand
  • 1,868
  • 90
  • PDF
The most dangerous code in the world: validating SSL certificates in non-browser software
TLDR
We present an in-depth study of SSL connection validation in non-browser software based on these APIs and present our recommendations. Expand
  • 417
  • 46
  • PDF
Airavat: Security and Privacy for MapReduce
TLDR
We present Airavat, a MapReduce-based system which provides strong security and privacy guarantees for distributed computations on sensitive data. Expand
  • 486
  • 40
  • PDF
How To Backdoor Federated Learning
TLDR
Federated learning enables thousands of participants to construct a deep learning model without sharing their private training data with each other. Expand
  • 327
  • 39
  • PDF
Exploiting Unintended Feature Leakage in Collaborative Learning
TLDR
We investigate passive and active property inference attacks that allow an adversarial participant in collaborative learning to infer properties of other participants’ training data that are not true of the class as a whole, but not generically for all class members. Expand
  • 302
  • 27
  • PDF
Constraint solving for bounded-process cryptographic protocol analysis
TLDR
We show how to convert the reachability problem into a constraint solving problem and present a relatively simple decision algorithm for the latter that is easy to understand and justify. Expand
  • 372
  • 27
  • PDF
Using Frankencerts for Automated Adversarial Testing of Certificate Validation in SSL/TLS Implementations
TLDR
We design, implement, and apply the first methodology for large-scale testing of certificate validation logic in SSL/TLS implementations. Expand
  • 143
  • 25
  • PDF
...
1
2
3
4
5
...