Portal de Periódicos da CAPES

A generic architecture for adaptive agents based on reinforcement learning

2003; Elsevier BV; Volume: 161; Issue: 1-2 Linguagem: Inglês

10.1016/j.ins.2003.03.005

1872-6291

Pierre‐Marie Preux, Samuel Delepoulle⋆, Jean‐Claude Darcheville,

Data Stream Mining Techniques

In this paper, we present MAABAC, a generic model for building adaptive agents: they learn new behaviors by interacting with their environment. These agents adapt their behavior by way of reinforcement learning, namely temporal difference methods. MAABAC is presented in its generality and then, different instantiations of the generic model are presented and experiments are reported. These experiments show the strength of this way of learning.