Definitions[]
A neural network is
“ | a software program whose operation imitates the way a human brain works. A neural network creates connections between processing elements, which are analogous to neurons. Given a certain pattern of input, the organization and weights of the connections determine the corresponding output. Most neural networks must be trained with a large set of "training data" before they can be effective in recognizing patterns within the input data. | ” |
“ | a weighted, directed graph where the vertices represent neurons and the edges represent connections (sometimes called synapses) between the neurons. The neurons are arranged in layers.[1] | ” |
Overview[]
"[N]eural networks, often described as being loosely modeled after the human brain, consist of thousands or millions of processing nodes generally organized into layers. The strength of the connections among nodes and layers are repeatedly tuned — based on characteristics of the training data — to correspond to the correct output. Advances in hardware, such as the development of graphical processing units (GPUs), have allowed these networks to have many layers, which is what puts the "deep" in deep learning. DL approaches have been used in systems across many areas of AI research, from autonomous vehicles to voice recognition technologies."[2]
References[]
See also[]
- Artificial neural network
- Cellular neural network
- Convolutional neural network
- Deep neural network
- Recursive neural network