[Colloquium] Sixiong Shan MS Presentation/Jul 22, 2022

[Megan Woodward]

This is an announcement of Sixiong Shan's MS Presentation
===============================================
Candidate: Sixiong Shan

Date: Friday, July 22, 2022

Time: 12:30 pm CST

Remote Location: https://urldefense.com/v3/__https://uchicago.zoom.us/j/6594501261?pwd=MVh4M3R4M0dPN1BaMUZBWDMrMnN1QT09__;!!BpyFHLRN4TMTrA!63006yW0f0Ujhnrtzw4zLb5X3KHJg-uynL8lc45sm-u7QIG96J9tSek83LURBr3FP2hojiA9uEVJdJ3QVvuS-8AyieXjp1G1vg$

Location: JCL 390

M.S. Paper Title: Poison Forensics: Traceback of Data Poisoning Attacks in Neural Networks

Abstract: In adversarial machine learning, new defenses against attacks on deep learning systems are routinely broken soon after their release by more powerful attacks. In this context, forensic tools can offer a valuable complement to existing defenses, by tracing back a successful attack to its root cause, and offering a path forward for mitigation to prevent similar attacks in the future. In this paper, we describe our efforts in developing a forensic traceback tool for poison attacks on deep neural networks. We propose a novel iterative clustering and pruning solution that trims "innocent" training samples, until all that remains is the set of poisoned data responsible for the attack. Our method clusters training samples based on their impact on model parameters, then uses an efficient data unlearning method to prune innocent clusters. We empirically demonstrate the efficacy of our system on three types of dirty-label (backdoor) poison attacks and three types of clean-label poison attacks, across domains of computer vision and malware classification. Our system achieves over 98.4% precision and 96.8% recall across all attacks. We also show that our system is robust against four anti-forensics measures specifically designed to attack it.

Advisors: Ben Zhao and Heather Zheng

Committee Members: Ben Zhao, Heather Zheng, and Rana Hanocka



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20220708/00c8d0e6/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: forensics.pdf
Type: application/pdf
Size: 609494 bytes
Desc: forensics.pdf
URL: <http://mailman.cs.uchicago.edu/pipermail/colloquium/attachments/20220708/00c8d0e6/attachment-0001.pdf>