A Dual-Mode Learning Mechanism Combining Knowledge-Education and Machine-Learning

Abstract

From 1956, the definitions of learning according to Artificial Intelligence and Psychology to human mind/behavior are obviously different. Owing to the rapid development of the computing power, we have potential to enhance the learning mechanism of AI. This work tries to discuss the learning process from the traditional AI learning models which are almost based on trial and error style. Furthermore, some relative literatures have pointed out that teaching-base education would increase the learning efficiency better than trial and error style. That is the reason we enhance the learning process to propose a dual-perspective learning mechanism, E&R-R XCS. As for XCS is a better accuracy model of AI, we have applied it as a basement and proposed to develop an intelligence-learning model. Finally, this work will give the inference discussion about the accuracy and accumulative performance of XCS, R-R XCS, and E&R-R XCS respectively, and the obvious summary would be concluded. That is, the proposed dual-learning mechanism has enhanced successfully.