• Amita Khatana
• V.K Narang
• Vikas Thada

In the field of computer engineering, image classification is emerging very rapidly for classifying images into different categories. A allot of approaches have been made to enhance the performance of image classification itself. Artificial neural network, support vector machine, decision tree are most popular approaches used for image classification. This paper focuses on one of the methods of image classification that is the deep artificial neural network. The deep artificial neural network consists of two or more hidden layers to extract the features of the image. The performance of artificial neural network depends upon its internal parameters such as weights, bias value, learning rate, batch size, number of neurons, hidden layers etc. To improve and update the internal parameters of the neural network, different optimization methods are used. The paper shows the comparative analysis of different deep learning optimization methods such as stochastic gradient descent,

Deep learning, ADAM, stochastic gradient descent, Adamax, Adadelta, image classification, MNIST

"COMPARATIVE ANALYSIS OF DEEP LEARNING OPTIMIZATION METHODS FOR IMAGE CLASSIFICATION ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 6, page no.513-521, June-2018, Available :http://www.jetir.org/papers/JETIR1806079.pdf

"COMPARATIVE ANALYSIS OF DEEP LEARNING OPTIMIZATION METHODS FOR IMAGE CLASSIFICATION ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 6, page no. pp513-521, June-2018, Available at : http://www.jetir.org/papers/JETIR1806079.pdf