Not much to say Aris, which I mean in the positive way offcourse because the less I say, the better it is
Great contribution

Cheers

Frazzle

I am writing a little test app that tries
to connect Neural Nets and a GA (that tries to optimize the net).

So the Neural Nets are actually not trained by a learning procedure
(like Backpropagation)
But evolved using a GA. (the evolution searches in the field of possible combinations
for the best sollution, and
the nets only act to their evolved properties without some magical
"backpropagation god")

This might not be the fastes choice, but the GA can go further and manipulate
the actual properties of the net too (connections, amount of neurons)

whats your opinion of this Idea:
A bot has 2 Neuronalnets:
The first net is always learning and copies itself after maybe 100 Inputs in the 2nd net, so the Bot is able to learn all the time?

The net that learns need to be the nets
thats processing the bots behaviour (or it could not learn).
So you only need one net after all.

But you could train a bot, and then save the properties of the net
for Bots in the published game.

Yes, but i thought a NN could just do one thing learning or working...

To learn, the net needs to work.
It can not learn by just observing (well not usually)

It needs to act (produce an output), and a "teacher" needs to adapt
its properties (weights, thresholds etc) according to the errors the net made
regarding the optimal output.
If a net is trained, it can be used to work on new data, and not learn anymore.

So a bot yould either use an already trained net, that does not change,
or even learn ingame furthermore.

Hello,

I can't find the file http://www.intenseillusions.com/GeneticAlgorithm.zip anymore. Anybody know where I could find it?

Thomas

Hey banjo,

The file is still there but my domain name changed. The new one is:

http://www.intense-i.com/GeneticAlgorithm.zip

ironically, i lost the old name to an illusionist/magician site

Thanks!

Thomas

This is a really a great contribution , I really dont understand any of it , but am very interested in figuring it out , thnx for the post and the link.