Understanding Smart Software: Two Useful Write Ups

July 2, 2015

I noted two articles about neural networks. The first purports to explain “Neural Networks in Plain English.” The article is useful and includes a useful hyperlink to information about genetic algorithms. The explanation requires the reader to have a basic grasp of mathematics and programming. Compared to other explanations of neural networks, this approach is pretty good.

The second document is “Neural Networks, Manifolds, and Topology.” It covers methods not directly addressed in the Plain English document; for example, some information about manifolds and a reference to k-nearest neighbor. In my lectures about the weaknesses of some text analytics systems, I discuss k-nearest neighbor, which is a versatile and interesting method.

Both write up warrant some attention. Keep in mind that the notions of “plain English” and “simple” are likely to carry a freight of meaning which can be surprising to some readers.

Stephen E Arnold, July 2, 2015

Comments

Comments are closed.

  • Archives

  • Recent Posts

  • Meta