A study on deep learning algorithms and architectures

Abstract

This review presented different types of deep architectures such as deep convolution networks, deep residual networks, recurrent neural networks, reinforcement learning, variational autoencoders etc. This study explained various deep neural networks, well-known training algorithms and architectures. This review also highlightedt heir shortcomings, e.g., getting stuck in the local minima, overfitting and training time for large problems ets