Title :
Towards associative non-algorithmic neural networks
Author :
Haikonen, Pentti O A
Author_Institution :
Lab. of Comput. & Inf. Sci., Helsinki Univ. of Technol., Espoo, Finland
fDate :
27 Jun-2 Jul 1994
Abstract :
A novel associative artificial neuron together with pulse train signals is proposed for the building block for non-algorithmic neural networks. These neurons operate on signal with inherent meaning and they may easily be cross connected and cascaded for various functions such as Pavlovian conditioning and concept/feature detectors. Short term and long term memory functions are also easily implemented and are in many circuits inherent
Keywords :
associative processing; learning (artificial intelligence); neural nets; pattern classification; Pavlovian conditioning; associative artificial neuron; associative nonalgorithmic neural networks; concept/feature detectors; long term memory functions; pulse train signals; short term memory functions; Artificial neural networks; Computer networks; Detectors; Electrostatic precipitators; Information science; Laboratories; Neural networks; Neurons; Signal processing; Temperature sensors;
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.1994.374270
