DocumentCode :
2286574
Title :
Pattern recognition using Boltzmann machine
Author :
Ma, Hede
Author_Institution :
Eng. Res. Labs., Savannah State Lab., GA, USA
fYear :
1995
fDate :
26-29 Mar 1995
Firstpage :
23
Lastpage :
29
Abstract :
An approach for pattern recognition using neural networks is proposed. Particularly, a Boltzmann machine, a Hopfield neural net model, is used in pattern recognition with desirable learning ability. The Boltzmann machine features stochastic learning, which acts as the connection dynamics for determining the weights on the connections between the neuron-like cells (processing elements) of different layers in the neural network. An algorithm for pattern recognition using Boltzmann machine is also presented, which could be coded with the C programming language or others to implement the approach for efficient pattern recognition
Keywords :
Boltzmann machines; C language; Hopfield neural nets; learning (artificial intelligence); pattern recognition; Boltzmann machine; C programming language; Hopfield neural net model; algorithm; connection dynamics; learning ability; neural networks; neuron-like cells; pattern recognition; processing elements; stochastic learning; Application software; Artificial intelligence; Containers; Humans; Image coding; Pattern analysis; Pattern recognition; Physics computing; Production systems; Real time systems;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Southeastcon '95. Visualize the Future., Proceedings., IEEE
Conference_Location :
Raleigh, NC
Print_ISBN :
0-7803-2642-3
Type :
conf
DOI :
10.1109/SECON.1995.513051
Filename :
513051
Link To Document :
بازگشت