Advanced Technologies for Industrial Applications by Rohit Thanki & Purva Joshi

Author:Rohit Thanki & Purva Joshi
Language: eng
Format: epub
ISBN: 9783031332388
Publisher: Springer International Publishing

5.1.4 Deep Learning

Deep learning is a subset of artificial intelligence and a newly developed learning method that uses neural networks [6] for output prediction. Nowadays, researchers in the literature propose many models based on this learning. In 1971, Dr. Robert Hecht-Nielsen, the inventor of the first neurocomputer, introduced the first neural network (NN) [6]. Artificial neural networks (ANN) are basically these networks. In his definition, a neural network is a computer system comprised of many simple, highly interconnected processing elements whose dynamic states respond to external inputs [6]. Application areas for this network include Big Data analysis, person recognition, and data prediction. The forwarding neural network (FNN) is also called this network. The network uses neurons arranged in tiers based on parallel operations. Figure 5.2 shows a simple artificial neural network. Input, hidden, and output layers are the main layers of the network. A neuron’s or node’s number depends on the size of its inputs and outputs. Nodes are fully connected to each other. Links with specific weighting values connect two nodes of adjacent layers.

Fig. 5.2Basic structure of artificial neural network

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.