Jigsaw Puzzle: Selective ML Backdoor Attack

Introduction

Further details can be found in our paper “Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifier” [PDF], IEEE Security and Privacy 2023.

If you end up building on this code or dataset as part of a project or publication, please include a reference to our paper:

@inproceedings{jigsawpuzzle,
  title = {Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifier},
  author={Yang, Limin and Chen, Zhi and Cortellazzi, Jacopo and Pendlebury, Feargus and Tu, Kevin and Pierazzi, Fabio and Cavallaro, Lorenzo and Wang, Gang},
  booktitle={2023 IEEE Symposium on Security and Privacy (SP)},
  year={2023}
}

Access

We are hosting Jigsaw Puzzle code on a private google drive folder, under open source license. To get access to the repository, please complete the following form:

> FORM TO REQUEST ACCESS <

Paper

PDF / Supplementary PDF / Slides

Talk

People

Limin Yang, Ph.D. from UIUC

Zhi Chen, Ph.D. student at UIUC

Jacopo Cortellazzi, Ph.D. student at King's College London

Feargus Pendlebury, Ph.D. from Royal Holloway, University of London

Kevin Tu, B.S. from UIUC

Fabio Pierazzi, Assistant Professor at King's College London

Lorenzo Cavallaro, Professor at University College London

Gang Wang, Associate Professor at UIUC