Dear Negin Malekian,

There are several works that are available at the attached links.

http://www.cs.cmu.edu/~mmv/papers/00TR-mike.pdf

http://link.springer.com/chapter/10.1007%2F978-3-642-27645-3_14#page-1

http://www.lirmm.fr/~jq/Cours/3cycle/module/HuWellman98icml.pdf

https://www.cs.duke.edu/courses/spring07/cps296.3/littman94markov.pdf

Thanks all,

Dear Lago,

What do you mean when you pose this:

"However, it is very difficult to model the coalescence, as you need to look at past generations."?

I think this paper (attached link) might help to understand the  relationship between evolutionary computing and reinforcement learning as it demonstrates the tie between learning automata and ant colony optimization.

In fact, it refers to ACO as colonies of learning automata in which every ant is modeled as simple unit of learning that help interconnected nodes to learn the optimal action.

http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=1049611