|
|
Computers > Artificial Intelligence > Genetic Programming > Ask the Community > Selecting Neural Net Parameters with Genetic Algorithms
Selecting Neural Net Parameters with Genetic Algorithms
Submitted 2005-11-08 11:55:36 by psiolent
Rating: 0 (0 votes) | 1143 Views
Section: Ask the Community
I have a project I'm about to undertake using a neural network to predict financial market fluctuations. The problem with such a data set is that there are literally thousands (perhaps tens of thousands) of potential parameters. Obviously any program that should run in a reasonable amount of time (especially real time) cannot process ALL of these parameters with neural network.
The key to a successful neural network thus becomes choosing the smallest parameter set with the most applicable data.
My question is this: has anyone had any experience using genetic algorithms to select a parameter set for use in a neural network? I've looked around a little for applicable research, but thought I would ask here to just to see if anyone could help.
Any advice, tips, cautions, etc. would be useful. |
Discuss this submission in the forum
|
|
|