notesum.ai

Published at November 29

Protecting Multiple Types of Privacy Simultaneously in EEG-based Brain-Computer Interfaces

cs.HC

cs.AI

cs.LG

Released Date: November 29, 2024

Authors: Lubin Meng, Xue Jiang, Tianwang Jia, Dongrui Wu

Arxiv: http://arxiv.org/pdf/2411.19498v1

Refer to caption

Privacy	Number of	Privacy	Original	Perturbed	Average
Type	Classes	Classifier	EEG	EEG	Reduction
Identity	54	EEGNet	25.55	4.52	20.03
		DeepCNN	34.98	3.26	31.72
		ShallowCNN	49.13	9.76	39.37
Gender	2	EEGNet	81.01	50.04	30.97
		DeepCNN	78.31	51.03	27.28
		ShallowCNN	82.31	51.24	31.07
BCI-	2	EEGNet	67.44	50.06	17.38
Experience		DeepCNN	70.17	51.09	19.08
		ShallowCNN	72.34	50.33	22.01
Average			62.25	35.70	26.55