In this study, we will observe how MNIST learning results vary depending on the depth of the artificial neural network, that is, the number of hidden layers, and analyze the correlation to describe it. The experiment was performed under the same conditions, with only the number of hidden layers adjusted. As a result of adjusting the hidden layer up to the 10th floor, it was found that a higher number of hidden layers does not necessarily lead to more accurate results.
Keywords
NND, MNIST, Neural Network Depth, NIST and AI