The objective of our research is to develop architectures for general intelligence. One mark of intelligence is rationality, that is, an intelligent agent (implemented in some architecture) uses its knowledge to control its own behavior to achieve its goals by deciding what problems to work on, what formulation of a problem to use, and what actions to take to solve a problem (Newell, 1982). The structure of the underlying architecture is critical in determining whether or not an agent is able to behave rationally and bring knowledge to bear when it is appropriate. The aim of this paper is to develop an architecture so that all aspects of problem-solving in an intelligent agent can be controlled by whatever knowledge is available to the agent.
KeywordsProblem Solver Intelligent Agent Problem Space Ancestor State Limited Rationality
Unable to display preview. Download preview PDF.