Publication Date:
2017
abstract:
In this paper, I describe the implications for a general theory of category formation
of a set of experiments in which embodied artificial agents are evolved for the ability to accomplish simple tasks. In particular, I will focus on how categories might
emerge from the dynamic interaction between an agent and its environment, and
on the relation between categories and behavior. Finally, I will introduce and discuss the notion of action-mediated categories, i.e., internal states that provide indirect and implicit information about the external environment and/or the agent/
environment relationship generated through the execution of a behavior.
Iris type:
02.01 Contributo in volume (Capitolo o Saggio)
Keywords:
category learning
List of contributors:
Nolfi, Stefano
Book title:
Handbook of Categorization in Cognitive Science