Difference between revisions of "History of Computers - Deep Learning"
(Created page with "Deep Learning ==Introduction== Deep learning is the next generation of <ref>History of Computers – Artificial Neural Networks|Neural Networks</ref>. “There are several ty...") |
(→Overview) |
||
(7 intermediate revisions by the same user not shown) | |||
Line 2: | Line 2: | ||
==Introduction== | ==Introduction== | ||
− | Deep | + | Deep Learning was invented by Geoffrey Hinton and is the next generation of [[History of Computers - Artificial Neural Network|Neural Networks]].<ref>https://ai.google/research/people/GeoffreyHinton</ref> “There are several types of Deep learning,deep neural networks, belief networks and recurrent networks.” <ref>https://www.educba.com/neural-networks-vs-deep-learning/</ref> |
==Overview== | ==Overview== | ||
http://neuralnetworksanddeeplearning.com/images/tikz41.png | http://neuralnetworksanddeeplearning.com/images/tikz41.png | ||
[http://neuralnetworksanddeeplearning.com/images/tikz41.png] | [http://neuralnetworksanddeeplearning.com/images/tikz41.png] | ||
+ | |||
While Neural Networks send data on a linear route through the hidden layer to the output layer, Deep Learning networks use multiple hidden layers to model high-level abstractions.<ref>http://neuralnetworksanddeeplearning.com/chap6.html</ref> Each layer uses the output from the previous layer as its input. Deep Learning networks. These networks can learn in one of two ways, either supervised – such as image recognition where inputs need to be classified – or unsupervised – where the computer is searching for a pattern. <ref>https://towardsdatascience.com/supervised-vs-unsupervised-learning-14f68e32ea8d</ref> | While Neural Networks send data on a linear route through the hidden layer to the output layer, Deep Learning networks use multiple hidden layers to model high-level abstractions.<ref>http://neuralnetworksanddeeplearning.com/chap6.html</ref> Each layer uses the output from the previous layer as its input. Deep Learning networks. These networks can learn in one of two ways, either supervised – such as image recognition where inputs need to be classified – or unsupervised – where the computer is searching for a pattern. <ref>https://towardsdatascience.com/supervised-vs-unsupervised-learning-14f68e32ea8d</ref> | ||
==Significance== | ==Significance== | ||
+ | Some of the notable uses of Deep Learning algorithms are drug synthesis, facial recognition, natural language processing and computer vision. <ref>https://www.deeplearningbook.org/contents/applications.html</ref> | ||
==Reference== | ==Reference== | ||
− | < | + | <references/> |
+ | Written by August Windham |
Latest revision as of 23:10, 17 September 2018
Deep Learning
Introduction
Deep Learning was invented by Geoffrey Hinton and is the next generation of Neural Networks.[1] “There are several types of Deep learning,deep neural networks, belief networks and recurrent networks.” [2]
Overview
While Neural Networks send data on a linear route through the hidden layer to the output layer, Deep Learning networks use multiple hidden layers to model high-level abstractions.[3] Each layer uses the output from the previous layer as its input. Deep Learning networks. These networks can learn in one of two ways, either supervised – such as image recognition where inputs need to be classified – or unsupervised – where the computer is searching for a pattern. [4]
Significance
Some of the notable uses of Deep Learning algorithms are drug synthesis, facial recognition, natural language processing and computer vision. [5]
Reference
- ↑ https://ai.google/research/people/GeoffreyHinton
- ↑ https://www.educba.com/neural-networks-vs-deep-learning/
- ↑ http://neuralnetworksanddeeplearning.com/chap6.html
- ↑ https://towardsdatascience.com/supervised-vs-unsupervised-learning-14f68e32ea8d
- ↑ https://www.deeplearningbook.org/contents/applications.html
Written by August Windham