Have nearly succeeded in building HyperNEAT but I'm having some trouble working out how to correct this error: [ 31%] Built target NEAT [ 87%] Built target...
Hey all, HyperNEAT C++ version 2.5 is out. This version has the following changes: -MPI support is working again and will compile successfully. -tinyxmldll...
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
Oct 4, 2008 7:35 am
4337
Hey Jason. Much delayed, I am finally trying to upgrade to 2.4....actually I'll just skip to 2.5 Quick question. I see you are now using Python for something....
The python module is a way to access hyperneat from python. It's not necessary to use hyperneat ... -- Sent from Gmail for mobile | mobile.google.com...
... *evolve* the ... Rechenberg's 1/5th ... heuristic. It ... fixed ... paper (cited ... global peak ... is on the ... helpful, ... CB shows ... rates, to ... ...
Peter, thanks for this HyperNEAT coevolution demo. I see some interesting behaviors evolving, but I am curious about your own conclusions? What did you learn...
Hey all, Per request, I added some flags to the cmake script that lets you customize whether or not to compile MPI & Python extensions. This reduces compile...
Hi Ken, I am happy that you like my demo. This particular experiment was set up to answer some questions I had in mind. The first one is obvious, is there a...
Peter, Thank you for the details. Is there any chance you could release a version that lets the user scale the networks and observe the impact? That would be...
Hi all, I've a question regarding adaptive and self-adaptive schemes and Jeff's paper: I understand that such approach could make things worse if the search...
Hi Jean-Baptiste, There is some research in this direction. I'm currently looking at Ofer Shir's recent PhD thesis, which combines niching with CMA-ES: ...
Yes, I am planning to upgrade the program so that the substrate configuration and other things will be loaded from a configuration file. That way, combined...
... That is a good question Ken. I think the problem is that evolution is short-sighted. It looks for what produces the highest average fitness in the short...
... Hello Jean-Baptiste- Thanks for your interest in the work. I think your question is an interesting one, but I do not think it affords an easy general...
Has anyone ever compared the performance of grammatical evolution to neuro evolution? Perhaps on the standard problems pole-balancing, xor. It seems quite...
Petar- I cannot get this to run on the two windows machines I have tried it on. Is there any way you could post a video of this as well as a video of your app ...
I understand that "hierarchical, modular, and formalized" are sometimes viewed as advantages for Grammatical Encoding (a similar argument is also made for...
I'm not sure that it's necessary to argue for GE from a purely encoding-based or search-space-based framework. It depends on what kind of attributes you value...
... And yet if evolution stumbles upon a module whose usefulness stems from multiple occurrences, how else would this be accomplished? If we want artificial...
Jean-Baptise, Thank you for your explanation. I understand your motivations and appreciate the idea. You are interested in being able to formalize encodings...
Thomas, you make a valid point. Some encodings are easier to decipher for humans than others. However, here is my guess: If an encoding is powerful enough...
... Derek, you probably know some of my thoughts on hierarchy and modularity from discussions we've had in person, but to summarize briefly, while I agree...
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.
