DarkFed: A Data-Free Backdoor Attack in Federated Learning
Abstract
We introduce First-Order Coalition Logic (FOCL), which combines key intuitions behind Coalition Logic (CL) and Strategy Logic (SL). Specifically, FOCL allows for arbitrary quantification over actions of agents. FOCL is interesting for several reasons. First, we show that FOCL is strictly more expressive than existing coalition logics. Second, we provide a sound and complete axiomatisation of FOCL, which, to the best of our knowledge, is the first axiomatisation of any variant of SL in the literature. Finally, while discussing the satisfiability problem for FOCL, we reopen the question of the recursive axiomatisability of SL.
Cite
Text
Li et al. "DarkFed: A Data-Free Backdoor Attack in Federated Learning." International Joint Conference on Artificial Intelligence, 2024. doi:10.24963/ijcai.2024/491Markdown
[Li et al. "DarkFed: A Data-Free Backdoor Attack in Federated Learning." International Joint Conference on Artificial Intelligence, 2024.](https://mlanthology.org/ijcai/2024/li2024ijcai-darkfed/) doi:10.24963/ijcai.2024/491BibTeX
@inproceedings{li2024ijcai-darkfed,
title = {{DarkFed: A Data-Free Backdoor Attack in Federated Learning}},
author = {Li, Minghui and Wan, Wei and Ning, Yuxuan and Hu, Shengshan and Xue, Lulu and Zhang, Leo Yu and Wang, Yichen},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2024},
pages = {4443-4451},
doi = {10.24963/ijcai.2024/491},
url = {https://mlanthology.org/ijcai/2024/li2024ijcai-darkfed/}
}