Title :
Analysis and pruning of temporally dynamic neural networks
Author_Institution :
Center for Autom. Res., Maryland Univ., College Park, MD, USA
fDate :
27 Jun-2 Jul 1994
Abstract :
An algorithm for pruning temporally dynamic neural networks is suggested. A set of “importance and similarity measures” are defined for both links and nodes of the network, which are computed recursively from output to input. Pruning and analysis of the network can be performed based on these importance and similarity measures. The suggested algorithm is tested on several networks including a “multi-rate temporal flow model” which is trained for a speaker independent phoneme recognition task
Keywords :
entropy; learning (artificial intelligence); neural nets; speech recognition; entropy; importance measures; multi-rate temporal flow model; nodes; pruning; similarity measures; speaker independent phoneme recognition; speech recognition; temporally dynamic neural networks; Automation; Computer networks; Educational institutions; Inspection; Neural networks; Performance analysis; Performance evaluation; Redundancy; Signal processing; Speech recognition;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374980
