Skip to main content

Advances, Systems and Applications

Table 3 The accuracy comparison of the method [17] and the optimized activation function method [18] with our method for analyzing The Cancer Genome Atlas (TCGA) data set is shown

From: Efficient and privacy-preserving image classification using homomorphic encryption and chunk-based convolutional neural network

Method

Dataset

Accuracy(%)

Activation Function

Multi-label Tumor

TCGA

85

softmax

Mish Fuction

Fashion-MNIST

84.68

Approximated Mish

Mish Fuction

CIFAR-10

70.2

Approximated Mish

Our Method

NEU-CLS

97.1

Square