Title :
Optimal artificial neural network architecture selection for bagging
Author :
Andersen, Tim ; Rimer, Mike ; Martinez, Tony
Author_Institution :
Iarchives, Provo, UT, USA
Abstract :
Studies the performance of standard architecture selection strategies, such as cost/performance and CV based strategies, for voting methods such as bagging. It is shown that standard architecture selection strategies are not optimal for voting methods and tend to underestimate the complexity of the optimal network architecture, since they only examine the performance of the network on an individual basis and do not consider the correlation between responses from multiple networks
Keywords :
Bayes methods; information theory; learning (artificial intelligence); neural net architecture; pattern classification; CV based strategy; bagging; cost/performance based strategy; optimal artificial neural network architecture selection; voting methods; Artificial neural networks; Bagging; Bayesian methods; Computer architecture; Costs; Diversity reception; Error correction; Neural networks; USA Councils; Voting;
Conference_Titel :
Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7044-9
DOI :
10.1109/IJCNN.2001.939460
