Jan 31, 2024 · We first present a general framework to comprehend the effect of poisoning attacks in FL when the training data is not independent and identically distributed.
Dec 9, 2024 · We first present a general framework to comprehend the effect of poisoning attacks in FL when the training data is not independent and identically distributed ...
Depriving the Survival Space of Adversaries Against Poisoned ...
dl.acm.org › doi › TIFS.2024.3360869
Jan 31, 2024 · We first present a general framework to comprehend the effect of poisoning attacks in FL when the training data is not independent and ...
This approach makes the local gradients less heterogeneous, allowing for the use of similarity detection techniques to filter out the poisoned gradients.
Depriving the Survival Space of Adversaries Against Poisoned Gradients in Federated Learning · J Lu · S Hu · W Wan · M Li · L Xue · H Wang · H Jin.
People also ask
What are the risks of federated learning?
What are the challenges of federated learning?
Depriving the Survival Space of Adversaries Against Poisoned Gradients in Federated Learning. Lu, J., Hu, S., Wan, W., Li, M., Zhang, L. Y., Xue, L., & Jin, H.
Cited by ; Depriving the Survival Space of Adversaries Against Poisoned Gradients in Federated Learning. J Lu, S Hu, W Wan, M Li, LY Zhang, L Xue, H Wang, H Jin.
合著作者 ; Depriving the Survival Space of Adversaries Against Poisoned Gradients in Federated Learning. J Lu, S Hu, W Wan, M Li, LY Zhang, L Xue, H Wang, H Jin.
Depriving the Survival Space of Adversaries Against Poisoned Gradients in Federated Learning. 2024IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY19 ...
Aug 20, 2024 · In this paper, we propose DPFLA, a novel scheme that can detect poisoning attacks without revealing the actual gradients of participants. DPFLA ...