Conference paper
Modifying the generalisation characteristics of a neural network with interactive reinforcement training
1997 IEEE International Conference on Intelligent Processing Systems (Cat. No.97TH8335), Vol.1, pp.472-476
IEEE
Proceedings of the 1997 IEEE International Conference on Intelligent Processing Systems, ICIPS'97 (Beijing, China, 28/10/1998–31/10/1998)
1998
Abstract
An interactive reinforcement training approach to modify the generalisation characteristics of a backpropagation neural network is proposed. The objective is to ensure that the network is capable of recognising significant training data even they are low in number. The interactive process will reinforce the important data by duplicating them. It ensures that the significant data are included in the final network. A case study of porosity prediction in petroleum exploration is used to illustrate this approach. Results have shown that the network's generalisation ability is modified to include the important outliners while avoiding overfitting. It is also useful in cases where training data are difficult or expensive to obtain.
Details
- Title
- Modifying the generalisation characteristics of a neural network with interactive reinforcement training
- Authors/Creators
- K.W. Wong (Author/Creator) - Curtin UniversityC.C. Fung (Author/Creator)H. Eren (Author/Creator)
- Publication Details
- 1997 IEEE International Conference on Intelligent Processing Systems (Cat. No.97TH8335), Vol.1, pp.472-476
- Conference
- Proceedings of the 1997 IEEE International Conference on Intelligent Processing Systems, ICIPS'97 (Beijing, China, 28/10/1998–31/10/1998)
- Publisher
- IEEE
- Identifiers
- 991005544665207891
- Copyright
- © IEEE 1998
- Murdoch Affiliation
- Murdoch University
- Language
- English
- Resource Type
- Conference paper
Metrics
200 File views/ downloads
66 Record Views