Title :
An algorithm for dynamically adapting neural network topologies
Author :
Piazza, F. ; Marchesi, M. ; Orlandi, G. ; Uncini, A.
Author_Institution :
Dipartimento di Elettronica ed Automatica, Ancona Univ., Italy
Abstract :
Recently, it has been proposed that biological networks change not only the synaptic strengths of connection but also partially their internal topologies, according to either the received external stimuli or the pre-existent connection layouts. Following this idea, a method is presented to dynamically adapt the topology of neural networks with supervised learning, using only the information of the training set. The method eliminates connections from an initial fully connected network, concurrently with the learning algorithm. Several experimental results obtained with multilayered networks are also reported to demonstrate the capabilities of the proposed method
Keywords :
learning systems; network topology; neural nets; biological networks; dynamic adaptation algorithm; internal topologies; learning algorithm; multilayered networks; neural network topologies; supervised learning; synaptic connection strength; training set; Artificial neural networks; Biological system modeling; Heuristic algorithms; Multilayer perceptrons; Network topology; Neural networks; Neurons; Optimal control; Size control; Supervised learning;
Conference_Titel :
Circuits and Systems, 1991. Conference Proceedings, China., 1991 International Conference on
Conference_Location :
Shenzhen
DOI :
10.1109/CICCAS.1991.184279