Multilayer Perceptron

This chapter discusses the working of artificial neural network (ANN) with single layer perceptron and multilayer perceptron along with example dataset. Also traditional linear regression and logistic regression are explained using python program. Along with these models (linear and logistic regression) description alternate approach have been explained using TensorFlow and Keras for better understanding of linear and logistic regression model. Each node in ANN has some weight and two functions for processing input. They are transfer function and activation function. A simple network with input layer and output layer (no hidden layer), which is considered as one of the first ANN used for binary classification task is called as single-layer perceptron. In the early stage, researchers are using regression model for classification. Later on they started using simple neural network with hidden layers in between input layer and output layer. This model is called MLP (Multilayer Perceptron).