SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2205.11819 · CVPR · 2022

Quarantine: Sparsity Can Uncover the Trojan Attack Trigger for Free

Tianlong Chen, Zhangyang Wang, Sijia Liu, Shiyu Chang, Zhenyu Zhang, Yihua Zhang

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 4 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
vita-group/backdoor-lth — 4 of 5
FunctionStatusWhere it lives
AverageMeter Ran vita-group/backdoor-lth/utils_reverse_trigger.py
code served (permissive licence) · get_code("58a51ee2ea69c0e6")
indent_str Ran vita-group/backdoor-lth/utils_reverse_trigger.py
code served (permissive licence) · get_code("f5a0090c6641566f")
remask Ran vita-group/backdoor-lth/utils_reverse_trigger.py
code served (permissive licence) · get_code("9545d137c9e673d5")
tanh_func Ran vita-group/backdoor-lth/utils_reverse_trigger.py
code served (permissive licence) · get_code("611b9f0f43be9479")
prints Not yet run vita-group/backdoor-lth/utils_reverse_trigger.py
code served (permissive licence) · get_code("d23d6d8a2f833624")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Trojan attacks threaten deep neural networks (DNNs) by poisoning them to behave normally on most samples, yet to produce manipulated results for inputs attached with a particular trigger. Several works attempt to detect whether a given DNN has been injected with a specific trigger during the training. In a parallel line of research, the lottery ticket hypothesis reveals the existence of sparse subnetworks which are capable of reaching competitive performance as the dense network after independent training. Connecting these two dots, we investigate the problem of Trojan DNN detection from the brand new lens of sparsity, even when no clean training data is available. Our crucial observation is that the Trojan features are significantly more stable to network pruning than benign features. Leveraging that, we propose a novel Trojan network detection regime: first locating a "winning Trojan lottery ticket" which preserves nearly full Trojan information yet only chance-level performance on clean inputs; then recovering the trigger embedded in this already isolated subnetwork. Extensive experiments on various datasets, i.e., CIFAR-10, CIFAR-100, and ImageNet, with different network architectures, i.e., VGG-16, ResNet-18, ResNet-20s, and DenseNet-100 demonstrate the effectiveness of our proposal. Codes are available at https://github.com/ VITA-Group/Backdoor-LTH.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2205.11819")
get_code_for_paper("2205.11819")
have("2205.11819")

Connect an agent — have() is free.