Genetic Algorithm

Reply #6

BuzzsawAnn — Sat, 15 Apr 2006 14:17:40 GMT

..... speaking of ants I bought this ant farm in this space age gel and it was marvelous to watch them digging holes to nowhere and piling this gel stuff on the side. Some ants were leaders and some fell by the wayside-some ants had a plan and some didn't know their heads from the hole in the gel. Last but not least it was always interesting to watch them move and bury the carcasses that got in the way of the digging.

The numbers for the day are 2951 2883 3690 2100

Reply #5

popeye — Wed, 15 Mar 2006 22:24:22 GMT

Ah yes, getting any prediction out of it is the hard part. Produces nice graphical effects but that's not of much use. I did work something out based on three objects (balls) being together and then being drawn to an attractor (the bottom of the mixing drum) and had some interesting results but I have not had time to test it out much. You have to have a good imagination with chaos and attractors!!!

Unfortunately, my math years were long ago and I have to work with very simple concepts.

I... [ More ]

Reply #4

bitrunner — Wed, 15 Mar 2006 20:06:57 GMT

How could particle optimization and ant optimization be used to say predict the sum of 5 or 6 balls? perhaps a fitness function employing some statistical function

Reply #3

popeye — Mon, 13 Mar 2006 22:13:25 GMT

Particle Swarm Optimization is similar to the ant colony theory. You can think of draw numbers being neighbours and try and find what other number those attract. There are C and VB source codes available.

Check out www.adaptiveview.com/articles/ipsop3.html

Reply #2

jordi marey — Mon, 13 Mar 2006 06:01:19 GMT

Gee Spot9 and Bitrunner

Maybe this GA product could backtest the Sat. Florida Lotto results for some of us here at LP.

Reply #1

Spot9 — Thu, 23 Feb 2006 01:27:06 GMT

Hi,

GA's are a way of searching a solution space in a few minutes, which could take years of computer time doing exhaustive search.

Another new method which is proving very fast is Ant Colony Optimisation

BBC

Ants might be able to run telecommunication networks better than humans.

Researchers have found that control programs based on the foraging behaviour of ants can keep data networks running more efficiently and cope with congestion better than many human alternatives.... [ More ]

Genetic Algorithm

bitrunner — Thu, 23 Feb 2006 00:45:32 GMT

Any one have good experience using Genetic Algorithm to select numbers?

Did you write your own or use commercial product