Ontogenetic Development of Skills, Strategies and Goals for Autonomously Behaving Systems

摘要

Biological organisms display an amazing ability during their ontogenetic development to adaptively develop solutions to the various problems of survival that their environments present to them. Dynamical and embodied models of cognition are beginning to offer new insights into how the numerous, heterogeneous elements of neural structures may self-organize during the development of the organism in order to effectively form adaptive categories and increasingly sophisticated skills, strategies and goals. The ontogenetic development of behavior in biological organisms represents a significant level of improvement over current approaches to machine learning. In this paper we discuss the possibility of building action selection mechanisms for autonomous agents based upon new insights into how exactly biological organisms manage to self-organize patterns of behavior during their ontogenetic development. We present a simple task environment that, nevertheless, affords opportunities for the hierarchical development of increasingly complex behaviors in humans. We present some results of standard machine learning mechanisms on performing this task. And finally we discuss future plans for developing models of the ontogenetic development of behavior for autonomous agents in the task environment.