Title :
From neurons to consciousness
Author_Institution :
Dept. of Comput. Sci., Austral Univ., Buenos Aires, Argentina
Abstract :
This paper outlines a possible model for human intelligence, based on the fact that the act of understanding can be explained, at best, as an act of pure intuition, impossible to explain in itself. Based on the model, we discuss how learning could be accomplished by such a system, and how knowledge could be transferred between generations. The model uses clusters of neural networks to achieve its goal. The proposed model uses a genetic algorithm closely coupled to the memory neural cluster to achieve creativity and to generate original responses. Other issues, such as possible internal representation models, are also discussed. As the model is certainly intensive in terms of computing-power and memory, and several components remain to be clearly defined, a reduced model is proposed. We define a system that can only handle one type of input (voice) that will learn a limited number of procedures. This reduced system would provide a basis to test the actual feasibility of the model and its possibilities, as well as different subsystem implementations
Keywords :
artificial intelligence; biocybernetics; genetic algorithms; learning systems; neural nets; creativity; genetic algorithm; human intelligence model; internal representation models; learning; memory neural cluster; neural networks; Artificial intelligence; Artificial neural networks; Computer science; Humans; Intelligent structures; Intelligent systems; Learning; Neurons; System testing; World Wide Web;
Conference_Titel :
Intelligence and Systems, 1996., IEEE International Joint Symposia on
Conference_Location :
Rockville, MD
Print_ISBN :
0-8186-7728-7
DOI :
10.1109/IJSIS.1996.565049