That could work. I'll try that tomorrow.

On another note though with stabilization, one of my colleagues took a look at a paper on Papov's theory for stability on fuzzy logic control systems, and followed it quite closely. We were hopping that the system would stabilize after 10,000 - 20,000 iterations; however, it actuality it was stable by time it had computed the second advertisement.

Screenshots in case you're curios: http://i.imgur.com/sjLwsnK.png

After 3k iterations, the system no longer had any noise. The only collapse that happened was at iteration 52k-ish.

Edit: I didn't realize I posted this from my personal account lol

I made various other optimizations to the point where calculating a score takes so little time that it isn't really worth it for me to implement this anymore