Over the past three decades or so, research on machine learning and data mining has led to
a wide variety of algorithms that learn general functions from experience. As machine …

Continual learning is the constant development of complex behaviors with no final end in
mind. It is the process of learning ever more complicated skills by building on those skills …

Reinforcement learning addresses the problem of learning to select actions in order to
maximize one's performance in unknown environments. To scale reinforcement learning to …

Action Selection schemes, when translated into precise algorithms, typically involve
considerable design effort and tuning of parameters. Little work has been done on solving …

This chapter introduces the major learning approach studied in this book: the explanation-
based neural network learning algorithm (EBNN). EBNN approaches the meta-level …

This paper shows how a single mechanism allows knowledge to be constructed layer by
layer directly from an agent's raw sensorimotor stream. This mechanism, the General Value …

Abstract Within Michigan-style Learning Classifier Systems based upon Holland's model
(Holland et al 1986) support for learning in delayed-reward multiplestep environments was …

The perception–action hierarchy contains a model of the environment as experienced based
on what has been recognized and done. Binons (binary neurons) can be used to represent …

Whilst the development of Learning Classifier Systems 1 has produced excellent results in
some fields of application, it has been widely noted that problems emerge when seeking to …

This dissertation presents a two level architecture for goal-directed robot control. The low
level actions are learned on-line as the robot performs its tasks, thereby reducing the need …