Multi-layer perceptron (MLP): Difference between revisions
| Line 21: | Line 21: | ||
* MLP's are fully connected (each hidden node is connected to each input node etc.) | * MLP's are fully connected (each hidden node is connected to each input node etc.) | ||
[[File:Mlp-network.png|thumb]] | [[File:Mlp-network.png|thumb]] | ||
Latest revision as of 21:47, 6 April 2018
This is student work which has not yet been approved as correct by the instructor
Case study notes[1]
Introduction[edit]
Multi-layer perceptrons are simply a type of neural network consisting of at least 3 nodes. The input nodes, the hidden nodes, and the output nodes. Each node is a neuron that 'activates' and turns on the next node etc.
<ref> https://en.wikipedia.org/wiki/Multilayer_perceptron</ref>
How does it work or a deeper look[edit]
- Multi-layer perceptrons use backpropagation as part of their learning phase.
- The nodes use a non-linear activation function (Basically they turn each other on)
- MLP's are fully connected (each hidden node is connected to each input node etc.)
External links[edit]
- It would be helpful
- to include many links
- to other internet resources
- to help fellow students
- Please make sure the content is good
- and don't link to a google search results, please