Title :
A multilayer feedforward neural network with adaptive lookup table weight
Author :
Dali, Yang ; Zemin, Lui
Author_Institution :
Beijing Univ. of Posts & Telecommun., China
Abstract :
A novel multilayer feedforward neural network model using the adaptive lookup table units as the neuron synapses and its learning algorithm are proposed. An improvement of the network model in performance over the conventional backpropagation (BP) network is the global convergence property. Also, the network shows much faster convergence speed as well as more time-saving iteration during the weight updating than the conventional feedforward network
Keywords :
convergence; feedforward neural nets; iterative methods; learning (artificial intelligence); table lookup; adaptive lookup table weight; convergence speed; global convergence property; learning algorithm; multilayer feedforward neural network; network model; neuron synapses; time-saving iteration; weight updating; Adaptive systems; Artificial neural networks; Convergence; Feedforward neural networks; Mathematics; Multi-layer neural network; Neural networks; Neurons; Switches; Table lookup;
Conference_Titel :
Circuits and Systems, 1993., ISCAS '93, 1993 IEEE International Symposium on
Conference_Location :
Chicago, IL
Print_ISBN :
0-7803-1281-3
DOI :
10.1109/ISCAS.1993.394250
