Title :
Pattern recognition using Boltzmann machine
Author_Institution :
Eng. Res. Labs., Savannah State Lab., GA, USA
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;
Conference_Titel :
Southeastcon '95. Visualize the Future., Proceedings., IEEE
Conference_Location :
Raleigh, NC
Print_ISBN :
0-7803-2642-3
DOI :
10.1109/SECON.1995.513051