Handwritten Digit Recognition

In recent times, with the proliferation of the Artificial Neural Network (ANN), deep learning has brought greater prominence to the field of machine learning by making it more intelligent. Deep learning is surprisingly used in a large range of fields due to a variety of applications such as surveillance, health, medicine, sports, robots, drones, etc. In deep learning, the Convolutional Neural Network (CNN) is at the center of amazing progress that integrates the Artificial Neural Network (ANN) with deep learning strategies to date. They are widely used in pattern recognition, sentence segmentation, speech recognition, facial recognition, text input, text analysis, incident, and handwritten digit recognition. The goal of this paper is to look at the difference in accuracy of CNN to separate handwritten numbers using different numbers of hidden layers and epochs and to make comparisons between accuracy. For this CNN performance evaluation, we performed our experiments using the Modified National Institute of Standards and Technology (MNIST) dataset. In addition, the network is trained using the stochastic gradient origin and the backpropagation algorithm