Practical training framework for fitting a function and its derivatives. Academic Article

abstract

• This paper describes a practical framework for using multilayer feedforward neural networks to simultaneously fit both a function and its first derivatives. This framework involves two steps. The first step is to train the network to optimize a performance index, which includes both the error in fitting the function and the error in fitting the derivatives. The second step is to prune the network by removing neurons that cause overfitting and then to retrain it. This paper describes two novel types of overfitting that are only observed when simultaneously fitting both a function and its first derivatives. A new pruning algorithm is proposed to eliminate these types of overfitting. Experimental results show that the pruning algorithm successfully eliminates the overfitting and produces the smoothest responses and the best generalization among all the training algorithms that we have tested.

authors

• Bukkapatnam, Satish

published proceedings

• IEEE Trans Neural Netw

author list (cited authors)

• Pukrittayakamee, A., Hagan, M., Raff, L., Bukkapatnam, S., & Komanduri, R.

citation count

• 14

complete list of authors

• Pukrittayakamee, Arjpolson||Hagan, Martin||Raff, Lionel||Bukkapatnam, Satish TS||Komanduri, Ranga

publication date

• June 2011

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

published in

• IEEE Transactions on Neural Networks and Learning Systems  Journal

keywords

• Algorithms
• Artificial Intelligence
• Computer Simulation
• Image Interpretation, Computer-Assisted
• Models, Theoretical
• Numerical Analysis, Computer-assisted
• Pattern Recognition, Automated

Digital Object Identifier (DOI)

• 10.1109/TNN.2011.2128344

start page

• 936

end page

• 947

volume

• 22

issue

• 6

URL

• http://dx.doi.org/10.1109/tnn.2011.2128344