Information concepts in anticipatory systems

Abstract

Anticipatory Systems involve intelligent information acquisition and processing. This presentation provides a combined information, communication, and computational perspective of these systems using an Information Engine model that leads to some fundamental definitions of intelligence. An Information Engine model represents the transformation of raw information to a form that is directly utilized by the target application just as a thermodynamic engine converts heat into mechanical work. Taking the analogy of Carnot's cycle, the area of the information cycle in the information-need and entropy coordinates of the Information Engine model is defined as logical work which is proposed here as a unified measure of intelligence that has the promise of capturing a variety of diverse systems ranging from natural to constructed and hybrid systems. This model provides a unified concept of the informational, computational, and intelligence aspects of anticipatory systems and hybrid intelligent systems across diverse implementations and applications.