Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Jamie Hayes

Unlocking High-Accuracy Differentially Private Image Classification through Scale


Apr 28, 2022
Soham De, Leonard Berrada, Jamie Hayes, Samuel L. Smith, Borja Balle


  Access Paper or Ask Questions

Reconstructing Training Data with Informed Adversaries


Jan 13, 2022
Borja Balle, Giovanni Cherubin, Jamie Hayes


  Access Paper or Ask Questions

Learning to be adversarially robust and differentially private


Jan 06, 2022
Jamie Hayes, Borja Balle, M. Pawan Kumar

* Preliminary work appeared at PPML 2021 

  Access Paper or Ask Questions

Towards transformation-resilient provenance detection of digital media


Nov 14, 2020
Jamie Hayes, Krishnamurthy, Dvijotham, Yutian Chen, Sander Dieleman, Pushmeet Kohli, Norman Casagrande


  Access Paper or Ask Questions

Adaptive Traffic Fingerprinting: Large-scale Inference under Realistic Assumptions


Oct 19, 2020
Vasilios Mavroudis, Jamie Hayes


  Access Paper or Ask Questions

Toward Robustness and Privacy in Federated Learning: Experimenting with Local and Central Differential Privacy


Sep 08, 2020
Mohammad Naseri, Jamie Hayes, Emiliano De Cristofaro


  Access Paper or Ask Questions

Provable trade-offs between private & robust machine learning


Jun 08, 2020
Jamie Hayes


  Access Paper or Ask Questions

Extensions and limitations of randomized smoothing for robustness guarantees


Jun 07, 2020
Jamie Hayes

* CVPR 2020 Workshop on Adversarial Machine Learning in Computer Vision 

  Access Paper or Ask Questions

Unique properties of adversarially trained linear classifiers on Gaussian data


Jun 06, 2020
Jamie Hayes


  Access Paper or Ask Questions

Contamination Attacks and Mitigation in Multi-Party Machine Learning


Jan 08, 2019
Jamie Hayes, Olga Ohrimenko


  Access Paper or Ask Questions

A note on hyperparameters in black-box adversarial examples


Nov 15, 2018
Jamie Hayes


  Access Paper or Ask Questions

Evading classifiers in discrete domains with provable optimality guarantees


Oct 25, 2018
Bogdan Kulynych, Jamie Hayes, Nikita Samarin, Carmela Troncoso


  Access Paper or Ask Questions

LOGAN: Membership Inference Attacks Against Generative Models


Aug 21, 2018
Jamie Hayes, Luca Melis, George Danezis, Emiliano De Cristofaro

* Proceedings on Privacy Enhancing Technologies (PoPETs), Vol. 2019, Issue 1 

  Access Paper or Ask Questions

Learning Universal Adversarial Perturbations with Generative Models


Jan 05, 2018
Jamie Hayes, George Danezis


  Access Paper or Ask Questions

Generating Steganographic Images via Adversarial Training


Jul 24, 2017
Jamie Hayes, George Danezis

* 9 pages 

  Access Paper or Ask Questions