On the Random Conjugate Kernel and Neural Tangent Kernel
Zhengmian Hu 1 Heng Huang 1
Abstract due to the difficulties raised by the non-convexity of the loss
function and the complication of optimization methods.
We investigate the distributions of Conjugate Ker-
nel (CK) and Neural Tangent Kernel (NTK) for One way to simplify the DNNs analysis is to only train
ReLU networks with random init ...


雷达卡




京公网安备 11010802022788号







