Abstract
Self-adaptive behavior can be defined as the behavior that allows an agent to adapt to a context using her/his/its resources. The property of being ‘self-adaptive’ implies considering some preliminary sources or elicitors for such skill. In the case of machine learning, all the learning or self-adaptive behavior mechanisms are related to algorithmic models of mathematical nature, while in the case of humans more subtle neurochemical and symbolic processes (logical and linguistic) are present. The purpose of this paper is to offer a theoretical analysis of the basic mechanisms related to learning processes, always oriented towards the creation of artificial cognitive systems which can implement such bioinspired mechanisms. Parafunctionality is the key innovative concept we introduce for applying bioinspired cognition to machine learning exploring a real mechanism still unexplored.
Acknowledgments
This research has been funded by an ICREA Acadèmia Grant. I thank Isard Boix for his empirical support of my research. Not forget the set of precise reviewers who improved the strength of this paper. Any mistake must be placed under my responsibility.
Disclosure statement
No potential conflict of interest was reported by the author(s).