Moduleco
2 - Concepts of the multi-agent's dynamics
2.2.
Individual learning and cognitive hierarchy of the agent's behaviour (b)
Best Response Strategy (conditional to available information and computation ability)
Machine Learning (i.e. learning by self reinforcement, as Q learning)
Rules selection by Genetic Algorithms & Classifier Systems (i.e. Luigi Marengo, Murat Yidilzoglu, Thomas Vallee)
Modal Logic (set of possible worlds) - Learning in games
(Walliser)
Bayesian inference : Exploration-Exploitation Dilemma (X-Armed Bandit : Gintis, Rothchild, Bourgine-Leloup)
Viability Constraint
& Autonomous agent (Aubin, Varela -Bourgine)
Denis.Phan@enst-bretagne.fr
;
Antoine Beugnard@enst-bretagne.fr