Gap-Based Estimation: Choosing the Smoothing Parameters for Probabilistic and General Regression Neural Networks. Zhong, M.; Coggeshall, D.; Ghaneie, E.; Pope, T.; Rivera, M.; Georgiopoulos, M.; Anagnostopoulos, G. C.; Mollaghasemi, M.; and Richie, S. In Neural Networks, 2006. IJCNN '06. International Joint Conference on, pages 1870-1877, 2006.
doi  abstract   bibtex   
Probabilistic neural networks (PNN) and general regression neural networks (GRNN) represent the knowledge by a simple but interpretable model that approximates the optimal classifier/predictor in the sense of expected value of accuracy. This model requires an important preset smoothing parameter, which is usually chosen by cross-validation or clustering. In this paper, we demonstrate the difficulties of both these approaches, discuss the relationship between this parameter and some of the data statistics, and attempt to develop a fast approach to determine the optimal value of this parameter. Finally, through experimentation we show that our approach, referred to as a gap-based estimation approach, is superior to the compared approaches.
@InProceedings{Zhong2006,
  author    = {Zhong, M. and Coggeshall, D. and Ghaneie, E. and Pope, T. and Rivera, M. and Georgiopoulos, Michael and Anagnostopoulos, Georgios C. and Mollaghasemi, Mansooreh and Richie, S.},
  title     = {Gap-Based Estimation: Choosing the Smoothing Parameters for Probabilistic and General Regression Neural Networks},
  booktitle = {Neural Networks, 2006. IJCNN '06. International Joint Conference on},
  year      = {2006},
  pages     = {1870-1877},
  abstract  = {Probabilistic neural networks (PNN) and general regression neural
	networks (GRNN) represent the knowledge by a simple but interpretable
	model that approximates the optimal classifier/predictor in the sense
	of expected value of accuracy. This model requires an important preset
	smoothing parameter, which is usually chosen by cross-validation
	or clustering. In this paper, we demonstrate the difficulties of
	both these approaches, discuss the relationship between this parameter
	and some of the data statistics, and attempt to develop a fast approach
	to determine the optimal value of this parameter. Finally, through
	experimentation we show that our approach, referred to as a gap-based
	estimation approach, is superior to the compared approaches.},
  doi       = {10.1109/IJCNN.2006.246908},
  keywords  = {knowledge representation;neural nets;regression analysis;data statistics;gap-based estimation;general regression neural networks;knowledge representation;preset smoothing parameter;probabilistic neural networks;smoothing parameters;Computer science;Engineering management;Industrial engineering;Kernel;Linear discriminant analysis;Neural networks;Predictive models;Smoothing methods;Statistics;Training data},
}
Downloads: 0