Poisoning with Cerberus: Stealthy and Colluded Backdoor Attack against Federated Learning
Authors: Xiaoting Lyu, Yufei Han, Wei Wang, Jingkai Liu, Bin Wang, Jiqiang Liu, Xiangliang Zhang
AAAI 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our extensive study on 3 large-scale benchmark datasets and 13 mainstream defensive mechanisms confirms that Cerberus Poisoning raises a significantly severe threat to the integrity and security of federated learning practices, regardless of the flourish of robust Federated Learning methods. We evaluate the attack performance of the distributed backdoor attack methods using 3 benchmark datasets of different application scenarios. We implement all the involved algorithms using Py Torch on an Ubuntu workstation with NVIDIA 3090 GPUs. |
| Researcher Affiliation | Academia | 1Beijing Key Laboratory of Security and Privacy in Intelligent Transportation, Beijing Jiaotong University, China 2INRIA, France 3Zhejiang Key Laboratory of Multi-dimensional Perception Technology, Application and Cybersecurity, China 4University of Notre Dame, USA |
| Pseudocode | Yes | Algorithm 1: Cerberus Poisoning |
| Open Source Code | Yes | Our code can be found at the link 1. 1https://github.com/xtlyu/Cer P |
| Open Datasets | Yes | We evaluate Cer P on 3 large-scale benchmark datasets: the applications of image classification (CIFAR-100 (Krizhevsky, Hinton et al. 2009) and Fashion-MNIST (Xiao, Rasul, and Vollgraf 2017)), and the loan/credit risk assessment (LOAN (George 2020)). |
| Dataset Splits | No | The paper discusses "training iterations" and "testing data" but does not explicitly provide information on train/validation/test dataset splits or their sizes/percentages, nor does it explicitly mention a validation set. |
| Hardware Specification | Yes | We implement all the involved algorithms using Py Torch on an Ubuntu workstation with NVIDIA 3090 GPUs. |
| Software Dependencies | No | We implement all the involved algorithms using Py Torch on an Ubuntu workstation with NVIDIA 3090 GPUs. The paper mentions PyTorch but does not specify its version number or versions for other software dependencies. |
| Experiment Setup | Yes | The datasets, hyperparameters, and model structures are summarized in Table 1. |