Please note: This PhD seminar will take place in DC 1304 and virtually.
Yiwei Lu, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Yaoliang Yu
Indiscriminate data poisoning attacks aim to decrease a model’s test accuracy by injecting a small amount of corrupted training data. Despite significant interest, existing attacks remain relatively ineffective against modern machine learning (ML) architectures.
In this work, we introduce the notion of model poisoning reachability as a technical tool to explore the intrinsic limits of data poisoning attacks towards target parameters (i.e., model-targeted attacks). We derive an easily computable threshold to establish and quantify a surprising phase transition phenomenon among popular ML models: data poisoning attacks can achieve certain target parameters only when the poisoning ratio exceeds our threshold. Building on existing parameter corruption attacks and refining the Gradient Canceling attack, we perform extensive experiments to confirm our theoretical findings, test the predictability of our transition threshold, and significantly improve existing indiscriminate data poisoning baselines over a range of datasets and models. Our work highlights the critical role played by the poisoning ratio, and sheds new insights on existing empirical results, attacks and mitigation strategies in data poisoning.
- Download the paper at https://arxiv.org/abs/2303.03592.
To join this PhD seminar in person, please go to DC 1304. You can also attend virtually using Zoom at https://uwaterloo.zoom.us/j/91737486953.
200 University Avenue West
Waterloo, ON N2L 3G1