Title :
Associative processor with memory based on neuron-like elements capable of temporal summation of input signals: processor for structural processing of information
Author :
Kharlamov, A. ; Raevsky, V.
Author_Institution :
Inst. of Higher Nervous Activity & Neurophysiol., Acad. of Sci., Moscow, Russia
Abstract :
In the report a model of associative processor with memory (APM) is considered. APM is a structural unit of the network consisting of neuron-like elements (NE) with temporal summation of input signals. APM realizes associative mapping of input information into signal space. This representation preserves the topology of perceptual space. On the basis of APM it is possible to build hierarchical structure for structural processing of the information.
Keywords :
associative processing; content-addressable storage; data structures; neural nets; signal processing; APM; associative processor; hierarchical structure; input signal temporal summation; neural memory; neuron-like elements; perceptual space topology preservation; structural information processor; Analytical models; Image recognition; Mathematical model; Mathematics; Network topology; Neurons; Neurophysiology; Registers; Signal mapping; Signal processing;
Conference_Titel :
Neural Information Processing, 2002. ICONIP '02. Proceedings of the 9th International Conference on
Print_ISBN :
981-04-7524-1
DOI :
10.1109/ICONIP.2002.1198970
