New project, Travelling salesman problem[need help]

The Partridge Family were neither partridges nor a family. Discuss.
User avatar
viruskiller
Posts: 399
Joined: June 14th, 2012, 5:07 pm

Re: New project, Travelling salesman problem[need help]

Post by viruskiller » August 2nd, 2012, 2:40 pm

i know what ants are lol:)).ive seen freeman's post on this thread and tought it might be something else called ants ,like a coding concept or something.

Freeman
Posts: 11
Joined: July 31st, 2012, 11:36 am

Re: New project, Travelling salesman problem[need help]

Post by Freeman » August 2nd, 2012, 6:40 pm

http://en.wikipedia.org/wiki/Ant_colony_optimization

here ya go virus killer, happy reading in one of the most awesome ideas in computer sience imho
( edit ) not that i know all of them, heh

User avatar
Asimov
Posts: 814
Joined: May 19th, 2012, 11:38 pm

Re: New project, Travelling salesman problem[need help]

Post by Asimov » August 2nd, 2012, 7:41 pm

Hi Freeman,

Interesting about ants on wikipedia.

From the page I have learnt that ants are basically lazy and if they can take a short cut to food they will. No different to humans heh heh.

Asimov
----> Asimov
"You know no matter how much I think I have learnt. I always end up hitting brick walls"
http://www.asimoventerprises.co.uk

User avatar
viruskiller
Posts: 399
Joined: June 14th, 2012, 5:07 pm

Re: New project, Travelling salesman problem[need help]

Post by viruskiller » August 2nd, 2012, 9:44 pm

now that ive seen the page i remember reading it once,i just completly forgot about it:D
ty for bring it up:)

Freeman
Posts: 11
Joined: July 31st, 2012, 11:36 am

Re: New project, Travelling salesman problem[need help]

Post by Freeman » August 2nd, 2012, 10:09 pm

well its more than just that, we did it at uni by assinging random very short code snippets to an ant colony ( pre-made obviously ) and then using criteria called fitness of each population you take the colonies with the strongest code and combine them using crossover methods ( so half of the one colonies code with half of the other colonies code ) and you keep doing this iteratively until you "ant running program" ( the objective that you set your ant to accomplish ) has effectively written itself. we didnt go much more indepth than that. for the german speakers amongst us i could link the lecture notes ( presentation style ): http://asv.informatik.uni-leipzig.de/up ... ADS213.pdf

Freeman
Posts: 11
Joined: July 31st, 2012, 11:36 am

Re: New project, Travelling salesman problem[need help]

Post by Freeman » August 2nd, 2012, 10:10 pm

the stuff with emergence of 'swarm behaviour' and 'intelligence' is where it starts getting really interesting, but i havent yet had the oportunity to go into much more detail than that so far

Post Reply