ABSTRACT
An artificial neural network (ANN) can be defined as a computational model based on the structure and functions of biological neural networks and find the complex relationship between input and output. The first ANN was invented in 1958 by a psychologist named Frank Rosenblatt. He coined the term as ‘perceptron’ and visualized it to be like the functionalities of a human brain. It was intended to model how the human brain processed visual data and learnt to recognize objects.
This chapter aims at introducing the building blocks of ANNs and its real-life applications. The content majorly focuses on the structure of biological neuron and how it inspired the discovery of ANNs.
The content of the chapter touches upon various forms of activation functions to decide which neuron should be fired by calculating a weighted sum of inputs and to introduce non-linearity to the output of a neuron. Various widely used backpropagation methods’ for optimizing the weights in the hypothesis. The final part of the chapter focuses on discussing the application of ANN in real-life scenarios, such as in healthcare, self-driving vehicles, agriculture, identifying the astronomical objects, etc.
