SOAR: An Architecture for General Intelligence

The ultimate goal of work in cognitive architecture is to provide the foundation for a system capable of general intelligent behavior -- i.e. to provide the underlying structure that would enable a system to perform the full range of cognitive tasks, employ the full range of problem-solving methods and representations appropriate to the tasks, and learn about all aspects of the tasks and its performance on them. This article presents Soar, an implemented proposal for such an architecture. We describe its organizational principles, the system as currently implemented, and demonstrations of its capabilities. Soar embodies 11 basic hypothesis about the structure of an architecture for general intelligence 1. Physical symbol-system--A general intelligence must be realized with a symbolic system, 2. Goal-structure-Control in a general intelligence is maintained by a symbolic goal system. 3. Uniform elementary- representation--There is a single elementary representation for declarative knowledge. Problem-space--Problem spaces are the fundamental organizational unit of all goal-directed behavior. 5. Production-system--Production systems are the appropriate organization for encoding all long-term knowledge. 6. Universal- subgoaling--Any decision can be an object of goal-oriented attention. 7. Automatic-subgoaling--All goals arise dynamically in response to impasses and are generated automatically by the architecture. 8. Control-knowledge--Any decision can be controlled by indefinite amounts of knowledge, both domain dependent and independent. 9. Weak-method-The weak methods form the basic methods of intelligence.