... I think I'm going to use a moving average and moving std. deviation. So there'll be an initial set of data used to sample the mean and standard deviation...
NEAT really shouldn't be allowed to form recurrent connections to solve XOR. It's not that there is anything "wrong" or "cheating" about it- it's just that...
Hi everyone, I would like to know if there is some implementaion of rtNEAT for ANJI. Thanks in advance! -- Cícero Augusto - Bacharelando em Ciência da...
Derek and I haven't written one. But it's possible someone else has. - Philip ... -- Philip Tucker ptucker@... Derek and I haven't written one. But it's...
I asked Derek before subscribe to this list and he told me the same thing, so I expected that somebody else had coded this feature. (sorry for my bad english,...
I haven't heard about any rtNEAT implementations other than my own. Of course, I don't know everything that's been done, so it's possible someone out there has...
Hi At long last, and only a year after I promised Ken, NEAT4J is released. NEAT4J, not surprisingly, is my Java implenmentation of NEAT. Please have a look,...
Hey Matt, Did you code something like rtNeat? ... -- Cícero Augusto - Bacharelando em Ciência da Computação - UFBA ... "Any fool can write code that a...
Hi Ken That would be great, the more people that use it, the better it becomes. You can link to either the sourceforge page below or http://neat4j.org. Cheers...
Hi, OK so over the weekend I modified my experiments to use Z-Score inputs along with continuous means and standard deviations. I've been running an experiment...
... Well, it's a good start! ... lot ... Very good point. Another thing you need to deal with (and this is very tricky) is that there is an inherent bias the...
OK, I have now completed the online documentation which is now complete (apart from one bit which is under dev). I have also updated the doc archive on...
NEAT4J is now linked from my homepage: http://www.cs.ucf.edu/~kstanley/ I used the neat4j.org address. Thanks for sharing your code with the community. ken ...
I'd like to introduce myself since I'm basically doing the same thing you are and would like to interact with other idividuals about this. I've read a...
... I do have data for companies that have gone bust or have been bought so far this year, but yes right now for my current experiment I only use companies...
Hi Jeff, ... I have to admit to not having read any papers on the subject for some years. Largely on the basis that anyone with a profitable system will ...
... market ... money ... occurs ... This is a function of the "Program Trading" desks in the banks. That's program in the non-computer sense. The practice of...
There really aren't any studies I can personally attest to since I have really just begun this learning process myself. I will note that it appears to me...
Sort of off topic here but has anyone noticed that finance.yahoo.com has changed the adjusted closes of some companies around February 2006. I first downloaded...
Hi Aleks, Well the strange thing is that I got the data from yahoo in the first place also. Maybe yahoo changed the way they calculate the adj closes sometime ...
I guess we are discussing integrity of data here :). Either way, Yahoo is usually good with dividends (at least when I try to look those up). I don't see...
Hello, This email message is a notification to let you know that a file has been uploaded to the Files area of the neat group. File :...
neat@yahoogroups.com
Sep 13, 2006 11:40 pm
2748
Hi Jeff, ... I don't know much about speech recognition but I am aware of the basic concept of analysing the frequency spectrogram generated using fixed length...
Why not consider expanding the definition of the genome to include a designated portion which defines a shallow GP program tree, utilizing a limited number of...
I finally figured out where things were going south in my code and got it to work nicely for the various known tests (standard XOR, pole balancing) compariable...
Hello all, Philip and I are starting to work with unsupervised learning algorithms in the context of NEAT, and we had some questions about Hebbian learning in ...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
