Emotions, learning and control

Abstract

There is a popular belief that emotions are mysterious gut feelings, beyond mathematical analysis. This paper counters the popular belief by describing a mathematical technique incorporating emotions into intelligent control systems and establishes relationships between mathematical, psychological and philosophical analyses of emotions. Many control systems incorporate signals that are directly related to emotional signals. The author argues that adaptive systems have to incorporate emotional signals, and the understanding of the roles of emotions and concepts is important for the design of intelligent system architectures. The paper analyses the learning process, describes the mathematics of the learning instinct, and presents aesthetic emotions as signals related to the learning values of objects. The developed mathematical technique is practically useful for engineering designs.