History of Computers - Artificial Neural Network
Introduction
Artificial neural networks (ANNs) are a type of machine learning AI that loosely mirror biological neural networks. [1] ANNS are useful at solving problems that the creator dose not completely understand. ANNs are also useful when dealing with incomplete or crazy data. ANNs solve problems by adapting – or learning. ANNs learn through trial and error. Every training run is slightly different than the one before it. The training process is repeated until the network outputs the desired product. A major use ANN's is in Facial Recognition [2]
Overview
ANNs consist of input nodes, a hidden layer and an output layer. The simplified version of what happens is that each input layer feeds the data to the each node of the hidden layer which is assigned has a unique weight. Then each node outputs a value based on its weight and the data received. The output is then, depending on the complexity of the network, sent to either another input layer or to the output layer. The system 'learns' by altering the weight values in the hidden layer. [3] [4] I would suggest watching the cited [2] video.
Significance
ANNs greatly expand the capability of computers by allowing computers to surpass the abilities of those programing them.[5] Neural networks were part first generation of AI, they are simple and, with enough of training time, they can do amazing things.
References
- ↑ https://www.zdnet.com/article/what-is-deep-learning-everything-you-need-to-know/
- ↑ http://wiki.sjs.org/wiki/index.php/History_of_Computers_-_Facial_Recognition
- ↑ https://www.frontiersin.org/research-topics/4817/artificial-neural-networks-as-models-of-neural-information-processing
- ↑ https://www.youtube.com/watch?v=aircAruvnKk
- ↑ https://www.computerworld.com/article/2591759/app-development/artificial-neural-networks.html
Created by August Windham
