A Metacognitive Classifier Using a Hybrid ACT-R/Leabra Architecture

AAAI Publications, Workshops at the Twenty-Fifth AAAI Conference on Artificial Intelligence

Yury Vinokurov, Christian Lebiere, Seth Herd, Randall O'Reilly

Last modified: 2011-08-24

Abstract

We present a metacognitive classifier implemented within a hybrid architecture that combines the strengths of two existing, mature cognitive architectures: ACT-R and Leabra. The classification of a set of items into pre- viously seen and novel categories (TRAIN and TEST, respectively) is carried out in ACT-R using metacogni- tive signals supplied by Leabra. The resulting system performance is analyzed as a function of various archi- tectural parameters, and future directions of research are discussed.

Full Text: PDF