Intralayer-Connected Spiking Neural Network with Hybrid Training Using Backpropagation and Probabilistic Spike-Timing Dependent Plasticity

Table 4

Comparison of different models on the CIFAR10 dataset.

Models	Structure		Average accuracy (%)	Max accuracy (%)
ANN in Wu et al. [38]	Net3	—	—	90.49
Zhang and Peng [6]	Net3	5	—	91.41
Wang et al. [13]	Net3	10	91.03	91.11
Wu et al. [38]	Net3	400	—	90.53
Rathi et al. [39]	ResNet20	250	—	92.22
Rathi and Roy [40]	VGG16	5	—	92.70
Sengupta et al. [41]	VGG16	2500	—	91.55
Amiri et al. [37]	Brain-inspired structure	—	—	93.1
ISNN	Net3	5	91.40	91.45
ISNN-HY	Net3	5	91.43	91.47

The bold values in the first column are the names of the proposed models. The bold values in the last two columns are the best average and max accuracies.