Using habituation as a simple and fundamental learning mechanism in an embodied artificial agent

Abstract

Habituation, a non-associative learning widely observed across phylogeny, is fundamental for adaptation and, thus, survival of living organisms. This paper investigates the main characteristics of habituation in order to present three new computational models inspired by habituation. We develop these models as part of the Iterant Deformable Sensorimotor Medium (IDSM), a recently developed abstract model of behavior formation. The characteristics of these models are studied and analyzed. Our long-term objective is to research new unsupervised learning mechanisms for artificial learning agents.

Exploring Sensitization in the Context of Extending the Behavior of an Artificial Agent

“…inhibition of the response to a repeatedly presented stimulus) could also help prevent over-generalization by minimising the impact of the repetitions of low-intensity stimuli. We started to work on those non-associative learning ( [9]). Moreover the dual-process theory (see [10]) and the behavioral homeostasis theory (see [8,6]) are works studying how non-associative learning fit together and what could be the impacts of such association.…”