... I hadn't actually read the "Automobile Warning System" portion of your dissertation, but I did so after Reuben's mention of it. It's interesting stuff....
The first thing I should clarify is that when my dissertation (and the chapter on RARS) was written, we had *not* used raw visual input, as you quoted....
On Thu, 02 Dec 2004 02:51:53 -0000, Kenneth Stanley ... Ah...okay. Mystery solved. ... I guess my next question is how you are feeding pixel input from the ...
Hi Derek, Yes, the pole-balancing approach is a proof-of-concept. I'm only going to be using 1 pole. I want to see whether training on high rez video is ...
Hi, I'm a beginning research student at University of Essex, UK (more on me at http://julian.togelius.com) and I'm also new to this group. My idea of the...
Hi Julian, I wrote Modular NEAT for my undergraduate thesis at UT; so, the GECCO paper was kind of a distillation of the thesis, with a little extra work on...
I'm trying to run JNeat on OS X. I've followed all the directions in USAGE.txt and I get the following message: 22:45:41: error in generation.startNeat() ...
Hi Joseph, Thans for pointing to your thesis, which I have now read. Good stuff. It did clear up most of the details I was wondering about, but it would be...
Hello, I just ported the Single pole balancing code from the C++ version to JNeat. I'm using the pole1startgenes from the C++ version and the parameters I'm...
Thanks a lot Eric for the help. I will definitely contact Stephane.Doncieux@... soon :) Ken! I did not join the US universities that offered admission...
... to ... parameters ... fitness is ... double ... is ... the ... input to ... The activation should be the same, but you should check to make sure that the...
... the ... Just front end, non-active vision. Basically, we feed the view out the front window. ... visually ... and ... seem ... would ... True, the richer...
Hi Julian, My forthcoming position document on the work I want to perform over the next few years has a lot of relevance to modularity. It is not specifically...
On Mon, 06 Dec 2004 05:39:20 -0000, Kenneth Stanley ... How big a scene is that? How many pixels are we talking about? I would imagine that the front end...
... Ok, sorry for the delay. I've posted a tarball at: http://www.cs.utexas.edu/~joeraii/mneat.tar.bz2 I cleaned up the package a bit and wrote a (short)...
... want a ... tests, ... on up ... genomes. ... test, but ... in ... poor ... This sounds like a good idea. I will keep it in mind. You know, ultimately...
... What has been surprising to me is how this problem has come up while evolving for relatively simple problems, e.g. with prey capture the difficulty level...
Hey Guys, I'm using jNeat and I just modified the code so I could do pole-balacing. It works fine, and now I'm trying to use image input. Right now the input...
Reuben, I don't know if Ugo is still reading this group so he may have more technical insight into the memory usage since he wrote JNEAT, but here are a few...
Reuben, java -Xmx<some big number> could also do it. This sets the maximum heap size. Use java -X for further instructions. Julian On Mon, 13 Dec 2004 21:54:30...
Hello Reuben (I truly ask forgiveness for my bad English and the delay but I have approached to single Internet point in determines periods , not every day !) ...
... distribution. ... were ... better, ... while ... the ... start' ... always ... solution is ... And ... problem, take ... search ... problems ... Hopefuly...
... Following on from this it occurs to me that domains such as prey capture aren't all that different to the roving eye type domains, that is we have a grid...
... egocentric ... if ... don't ... the ... type of ... My roving eye actually was able to rotate by 90 degree increments. However, it could not zoom, as you...
Colin, my office mate and I were discussing phased vs. blended searching and we were looking at your graphs from your phased writeup and we realized something...
... Ah well, the problem here is that each line on the graph is an average of 3 runs. Therefore the fitness oscillations seen aren't indicative of the...
Thanks for your suggestions, Ken and Julian. I bumped it up to a gig with -Xmx1000m and it works now. Normal single pole-balancing works now with JNeat, and...
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.
