Join the initiative for modernizing math education.

Solve integrals with Wolfram|Alpha.
Photo provided by Flickr

Knowledge-based programming for everyone.

The original paper released by Teuvo Kohonen in 1998 consists on a brief,masterful description of the technique. In there, it is explained thata self-organizing map is described as an (usually two-dimensional) grid ofnodes, inspired in a neural network. Closely related to the map, is the idea ofthe model, that is, the real world observation the map is trying torepresent. The purpose of the technique is to represent the model with a lowernumber of dimensions, while maintaining the relations of similarity of thenodes contained in it.

Note (2018-02-01): You can also read , translated by YibingDu.
Photo provided by Flickr

In case you hadn’t noticed, I’m a bit of a stickler for terminology

To capture this similarity, the nodes in the map are spatially organized to becloser the more similar they are with each other. For that reason, SOM are agreat way for pattern visualization and organization of data. To obtain thisstructure, the map is applied a regression operation to modify the nodesposition in order update the nodes, one element from the model ((e)) at atime. The expression used for the regression is:

Aromatherapy - The Balance & Harmony of Body and …
Photo provided by Flickr

To solve it, we can try to apply a modification of the Self-Organizing Map(SOM) technique. Let us take a look at what this technique consists, and thenapply it to the TSP once we understand it better.

In mathematics, a function is a relation between a set of inputs and a set of permissible outputs with the property that each input is related to exactly one output.
Photo provided by Flickr