A study on the origin of anticipation by guidance for artificial dynamic cognition

Kristen Manac'H (1) , Pierre de Loor (1, 2)

1 LISYC - Laboratoire d'Informatique des Systèmes Complexes
2 CERV - Centre Européen de Réalité Virtuelle

This paper describes an evolutionary robotics experiment, which aims at showing the possibility of anticipation and learning in an enaction-inspired artificial intelligence perspective. After a theoretical positioning, we present the experiment and the first results.

Mots clés: Continuous; Time; Reccurent; Neural; Networks; Hebbian; plasticit; Enactive Artificial Intelligence; Reinforcement Learning

Domaines: Informatique; Intelligence artificielle [cs.AI]

Kristen Manac'H, Pierre de Loor. A study on the origin of anticipation by guidance for artificial dynamic cognition. Fourth workshop on: Anticipatory Behaviour in Adaptive Learning Systems (ABiALS 2008), 2008, France. ⟨hal-00611021⟩