On Using Constructivism in Neural Classifier Systems

For artificial entities to achieve true autonomy and display complex life-like behaviour they will need to exploit appropriate adaptable learning algorithms. In this sense adaptability implies flexibility guided by the environment at any given time and an open-ended ability to learn novel behaviours. This paper explores the potential of using constructivism within the neural classifier system architecture as an approach to realise such behaviour. The system uses a rule structure in which each is represented by an artificial neural network. Results are presented which suggest it is possible to allow appropriate internal rule complexity to emerge during learning and that the structure indicates underlying features of the task.