How many hidden layers and nodes?
Top Cited Papers
- 20 April 2009
- journal article
- research article
- Published by Informa UK Limited in International Journal of Remote Sensing
- Vol. 30 (8), 2133-2147
- https://doi.org/10.1080/01431160802549278
Abstract
The question of how many hidden layers and how many hidden nodes should there be always comes up in any classification task of remotely sensed data using neural networks. Until today there has been no exact solution. A method of shedding some light to this question is presented in this paper. A near‐optimal solution is discovered after searching with a genetic algorithm. A novel fitness function is introduced that concurrently seeks for the most accurate and compact solution. The proposed method is thoroughly compared to many other methods currently in use, including several heuristics and pruning algorithms. The results are encouraging, indicating that it is time to shift our focus from suboptimal practices to efficient search methods, to tune the parameters of neural networks.Keywords
This publication has 19 references indexed in Scilit:
- The application of artificial neural networks to the analysis of remotely sensed dataInternational Journal of Remote Sensing, 2007
- Learning capability and storage capacity of two-hidden-layer feedforward networksIEEE Transactions on Neural Networks, 2003
- Evolutionary ensembles with negative correlation learningIEEE Transactions on Evolutionary Computation, 2000
- Pruning artificial neural networks: An example using land cover classification of multi-sensor imagesInternational Journal of Remote Sensing, 1999
- Strategies and best practice for neural network image classificationInternational Journal of Remote Sensing, 1997
- The Use of Artificial Neural Networks in a Geographical Information System for Agricultural Land-Suitability AssessmentEnvironment and Planning A: Economy and Space, 1994
- Training feedforward networks with the Marquardt algorithmIEEE Transactions on Neural Networks, 1994
- Forecasting with neural networksInformation & Management, 1993
- Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing DataIEEE Transactions on Geoscience and Remote Sensing, 1990
- A note on genetic algorithms for large-scale feature selectionPattern Recognition Letters, 1989