Self-awareness and Self-control in NARS

This paper describes the self-awareness and self-control mechanisms of a general-purpose intelligent system, NARS. The system perceives its internal environment basically in the same way as how it perceives its external environment, though the sensors involved are completely different. NARS uses a “self” concept to organize its relevant beliefs, tasks, and operations. The concept has an innate core, though its content and structure are mostly acquired gradually from the system’s experience. The “self” concept and its ingredients play important roles in the control of the system. Functions like “self-awareness”, “self-control”, and “self-consciousness” are closely related to advanced forms of intelligence. The difficulty of realizing these functions in a machine is both technical and theoretical, as there is no widely accepted theory about them, and even their definitions are highly controversial. This paper is not an attempt to address all relevant issues. Instead, we will present the relevant aspects of NARS (Non-Axiomatic Reasoning System), a formal model of general intelligence, which has been mostly implemented and is under testing and tuning. In the following, the conceptual design of NARS is briefly introduced first, then the parts mostly relevant to “self” are described in more detail. Finally, the major design decisions are compared with the related works.