Title :
CAM-based ASOCS implementation
Author :
Bartczak, Andrew ; Daly, James
Author_Institution :
Dept. of Electr. Eng., Rhode Island Univ., Kingston, RI, USA
fDate :
27 Jun-2 Jul 1994
Abstract :
We describe an implementation of adaptive self-organizing concurrent system (ASOCS) using content addressable memory (CAM). ASOCS is an artificial neural network with supervised online learning trained by incrementally introduced Boolean propositional logic rules or instances. Words stored in CAM represent modified instances after they are processed by ASOCS to ensure consistency and minimality of the entire collection of instances. CAM permits fast training and execution. Main advantages of our implementation are simplicity, scalability to handle wider input and output vectors, ability to preset the network structure in advance based on software simulation
Keywords :
Boolean functions; adaptive systems; content-addressable storage; learning (artificial intelligence); neural chips; neural nets; Boolean propositional logic rules; adaptive self-organizing concurrent system; content addressable memory; input vectors; neural network; output vectors; scalability; supervised online learning; Adaptive systems; Artificial neural networks; Associative memory; Boolean functions; CADCAM; Computer aided manufacturing; Hardware; Propagation delay; Scalability; Very large scale integration;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374539
