Newton methods based convolution neural networks using parallel processing

Ujjwal Thakur, Anuj Sharma

Training of convolutional neural networks is a high dimensional and a non-convex optimization problem. At present, it is inefficient in situations where parametric learning rates c…