Neurocomputational theories of homeostatic control
Résumé
Homeostasis is a problem for all living agents. It entails predictively regulating internal states within the bounds compatible with survival in order to maximise fitness. This can be achieved physiologically, through complex hierarchies of autonomic regulation, but it must also be achieved via behavioural control. Here we review some of the major theories of homeostatic control and their historical cognates, addressing how they tackle the optimisation of both physiological and behavioural homeostasis. We start with optimal control approaches, setting up key concepts, and expanding on their limitations. We then move onto contemporary approaches, in particularly focusing on a branch of reinforcement learning known as homeostatic reinforcement learning (HRL). We explain its main advantages, empirical applications, and conceptual insights. We then outline some challenges to HRL and reinforcement learning in general, and how survival constraints and Active Inference models could circumvent these problems.
Origine : Fichiers produits par l'(les) auteur(s)