🚀 go-pugleaf

Hello,

This is my first posting to cag, so you will have to excuse my ignorance
of basic terms. I am interested in applying various learning methods to
fuzzy logic systems and a very popular one seems to be genetic algorithms.

My intuition suggests that a genetic algorithm must be similar to a
directed random search. I realise it is not quite this simple, but I was 
wonderring what are the main differences. With all the talk on the net of
genes and mutations I assume that only part of the characterising 
information is altered with each iteration, but how in practice is this 
done and how do you combine genes.

Apologies for my ignorance, but I want to get an idea of what a genetic
algorithm is at a system level before trying to understand the detail.

Thanks for reading this,

Any answers to the net or e-mail to mdb1@eng.cam.ac.uk

Cheers,

Malcolm Binns.