Memory-Based Learning in Intelligent Control Systems

Memory-based learning is becoming more popular in Artificial Intelligence and can be used in Intelligent Control Systems. This talk will describe how an associative content-addressable memory can be used to model a robot and the world the robot interacts with. The model can be learned by storing experiences in the memory. To make predictions the memory is searched for relevant experience. An initial implementation of such a memory-based modeling scheme has been made on a parallel computer, the Connection Machine. The implementation was used to model and control a simulated planar two-joint arm and a simulated running machine. This paper describes the issues and problems that arose in this preliminary work (Atkeson and Reinkensmeyer 1989).