Marksman Backdoor: Backdoor Attacks with Arbitrary Target Class
Nov 28, 2022
Speakers
Khoa D. Doan
Speaker · 0 followers
Yingjie Lao
Speaker · 0 followers
About
In recent years, machine learning models have been shown to be vulnerable to backdoor attacks. Under such attacks, an adversary embeds a stealthy backdoor into the trained model such that the compromised models will behave normally on clean inputs but will misclassify according to the adversary's control on maliciously constructed input with a trigger. While these existing attacks are very effective, the adversary's capability is limited: given an input, these attacks can only cause the model to misclassify toward a single pre-defined or target class. In contrast, this paper exploits a novel backdoor attack with a much more powerful payload, denoted as Marksman, where the adversary can arbitrarily choose which target class the model will misclassify given any input during inference. To achieve this goal, we propose to represent the trigger function as a class-conditional generative model and to inject the backdoor in a constrained optimization framework, where the trigger function learns to generate an optimal trigger pattern to attack any target class at will while simultaneously embedding this generative backdoor into the trained model. Given the learned trigger-generation function, during inference, the adversary can specify an arbitrary backdoor attack target class, and an appropriate trigger causing the model to classify toward this target class is created accordingly. We show empirically that the proposed framework achieves high attack performance (e.g., 100% attack success rates in several experiments) while preserving the clean-data performance in several benchmark datasets, including MNIST, CIFAR10, GTSRB, and TinyImageNet. The proposed Marksman backdoor attack can also easily bypass existing backdoor defenses that were originally designed against backdoor attacks with a single target class. Our work takes another significant step toward understanding the extensive risks of backdoor attacks in practice.In recent years, machine learning models have been shown to be vulnerable to backdoor attacks. Under such attacks, an adversary embeds a stealthy backdoor into the trained model such that the compromised models will behave normally on clean inputs but will misclassify according to the adversary's control on maliciously constructed input with a trigger. While these existing attacks are very effective, the adversary's capability is limited: given an input, these attacks can only cause the model to…
Organizer
NeurIPS 2022
Account · 961 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
DeepFoids: Adaptive Bio-Inspired Fish Simulation with Deep Reinforcement Learning
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Panel Discussion: "Heading for a Unifying View on nCSI"
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Sustainable Online Reinforcement Learning for Auto-bidding
Watch later
Zhiyu Mou, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Predicting Immune Escape with Pretrained Protein Language Model Embeddings
Watch later
Kyle Swanson, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Is L2 Loss Always Suitable for Training Physics-Informed Neural Networks?
Watch later
Shanda Li, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Learning from uncertain concepts via test time interventions
Watch later
Ivaxi Sheth, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%