Table 5.

Setting temperature parameter $t=3$, the number of independent training epochs ${\mathbb{T}}_{in}=10$, change the number of distillation epochs ${\mathbb{T}}_d$, the accuracy (%) of cooperative training of two nodes (AlexNet + AlexNet) using our method.

Distillation Epochs (${\mathbb{T}}_{\mathit{d}}$)	1	5	10	20	30	40
AlexNet	77.72	78.29	78.9	78.84	79.43	79.41
AlexNet	77.96	78.54	78.96	79.04	79.59	79.31