Introduction

With the advancement of Deep Learning for image classification in 2012, Convolutional Neural Network extracted the image features and successfully classified the objects. In 1956, the Dartmouth Summer Research Project on Artificial Intelligence linked up artificial intelligence with neural network for joint project development. The neural network models are divided into supervised, semi‐supervised, and unsupervised learning. The software frameworks are developed to support deep learning applications. The most popular academic framework is Caffe which is migrated into Pytorch. Pytorch is a deep learning research platform that provides maximum flexibility and speed. Different neural network models are compared in terms of computational complexity, model efficiency, and memory utilization. To improve computational efficiency, new deep learning hardware architectures are developed to support intensive computation and high‐memory bandwidth demands.