Learning automata: an introduction
Learning automata: an introduction
Ant algorithms for discrete optimization
Artificial Life
Introduction to Reinforcement Learning
Introduction to Reinforcement Learning
Ant system: optimization by a colony of cooperating agents
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.00 |
In this paper we place ant algorithms in a reinforcement learning framework. We concentrate on the original Ant System and we briefly discuss Ant Colony system. We show that ant-quantity and ant-density can be considered as TD(0) algorithms which only take into account immediate reinforcement. Whereas ant cycle is basically an on-policy Monte Carlo method. We introduce the notion of decay traces, for modeling the decay of trail.