The NeuralBAG algorithm: optimizing generalization performance in bagged neural networks.
作者： John Carney , Padraig Cunningham
DOI:
关键词:
摘要: In this paper we propose an algorithm call \NeuralBAG" that estimates the set of weights and number hidden units each network in a bagged ensemble should have so generalization performance is optimized. Experiments performed on noisy synthetic data demonstrate potential algorithm. On average, ensembles trained using NeuralBAG out-perform networks cross-validation by 53% individual \cheating"
参考文章(5)
Leo Breiman, OUT-OF-BAG ESTIMATION ,(1996)
Robert J Tibshirani, Bradley Efron, An introduction to the bootstrap ,(1993)
Tom Heskes, Balancing Between Bagging and Bumping neural information processing systems. ,vol. 9, pp. 466- 472 ,(1996)
Anders Krogh, Richard G. Palmer, John Hertz, Introduction To The Theory Of Neural Computation ,(1991)
Leo Breiman, Bagging predictors Machine Learning archive. ,vol. 24, pp. 123- ,(1996) , 10.1023/A:1018054314350