ARTIFICIAL NEURAL NETWORKS
Need help with a related project topic or New topic? Send Us Your Topic
DOWNLOAD THE COMPLETE PROJECT MATERIAL
ARTIFICIAL NEURAL NETWORKS
Chapter one
1.1 Introduction
Bhadeshia H.(2019). Artificial neural networks are the simplest way to replicate the human brain, with neurons serving as the building pieces. There are around 100 billion neurons in the human brain.
Each neuron has a connecting point ranging between 1,000 and 100,000. knowledge is stored in the human brain in a scattered manner, and we can access many pieces of this knowledge from our memory at the same time if necessary.
We are not incorrect when we state that a human brain is composed of thousands of extremely strong parallel processors. Neurons in multi-layer artificial neural networks are similarly arranged in a manner comparable to that of the human brain.
Each neuron is coupled to other neurons using certain coefficients. During training, information is transferred to these connection points, causing the network to learn.The most advanced supercomputers.
Dr. Robert Hecht-Nielsen, the developer of one of the earliest neurocomputers, provides the most concise explanation of a neural network, sometimes known as a ‘artificial’ neural network (ANN).
He defines a neural network as a computing system comprised of a number of simple, highly interconnected processing components that process information based on their dynamic state responses to external inputs.
ANNs are processing devices (algorithms or hardware) that are roughly based on the neural organisation of the mammalian cerebral cortex, although on much smaller scale.
A big ANN may have hundreds or thousands of processor units, but a mammalian brain has billions of neurons, resulting in an increase in the size of overall interaction and emergent behaviour.
Although most ANN researchers are not concerned with whether their networks properly mirror biological systems, a few are. For example, researchers have successfully recreated retinal function and modelled the eye (Egmont-Petersen, 2020).
Artificial Neural Networks
1.2 Background of Artificial Neural Networks (ANNs)
An Artificial Neural Network (ANN) is a data processing paradigm inspired by how biological nervous systems, such as the brain, process information. The innovative structure of the information processing system is the defining feature of this paradigm.
It consists of a huge number of highly interconnected processing elements (neurons) that work together to solve specific challenges. ANNs, like humans, learn through example.
An artificial neural network (ANN) is trained for a specific application, such as pattern recognition or data classification. Learning in biological systems entails changes to the synaptic connections that occur between neurones. This applies to ANNs as well (Bishop and Christopher, 2015).
Artificial neural networks contain artificial neurons known as units. These units are organised into layers, which collectively make up the entire Artificial Neural Network in a system. The number of units in a layer can range from a dozen to millions, depending on the system’s complexity.
Artificial neural networks typically consist of an input layer, an output layer, and hidden layers. The input layer collects data from the outside world that the neural network must analyse or learn about.
The data is then passed through one or more hidden layers, which turn it into useful data for the output layer. Finally, the output layer produces an output in the form of an Artificial Neural Network’s reaction to the input data (Borgelt and Christian, 2018).
The majority of neural networks connect units from one layer to the next. Each of these links has a weight, which determines how much one unit influences another. As the data moves from one unit to the next, the neural network learns more about it, eventually producing an output from the output layer.
1.3 Advantages of the Artificial Neural Network (ANN)
Storing information over the network: Similar to traditional programming, information is stored across the network rather than in a database. The network continues to function even if a few pieces of information disappear in one location.
Ability to work with incomplete knowledge: Following ANN training, the data may produce output with incomplete information. The performance loss in this case is determined by the relevance of the missing information.
Being fault-tolerant: The corruption of one or more ANN cells does not prohibit the network from producing output. This feature makes the network fault-tolerant.
Having a distributed memory: In order for an ANN to learn, it must first determine the examples and then educate the network to produce the required output by displaying these examples to the network.
The network’s success is directly related to the selected examples, and if the event cannot be shown to the network in all of its dimensions, the network may produce erroneous output.
Gradual corruption: Over time, a network slows and degrades. The network problem does not appear to corrode immediately.
Machine learning capability: Artificial neural networks learn from and make conclusions based on similar experiences.
Parallel processing capability: Artificial neural networks have the numerical strength to accomplish several tasks simultaneously.
Need help with a related project topic or New topic? Send Us Your Topic