Training of Radial Basis Function Networks with H∞ Filter-Initial Simulation Results
Апстракт
This paper analyzes the application of the H∞ filter for the optimization of the parameters of an artificial neural network with Gaussian-type radial activation functions. The analysis showed that the H∞ filter generates better estimates of the parameters of the artificial neural network than the linearized Kalman filter in problems where there is significant initial parameter uncertainty, insufficient knowledge of system/process characteristics, process noise, and measurement noise. Unlike the linearized Kalman filter, the H∞ filter does not rely on the assumption that process and measurement noises are subject to Gaussian distribution, which is a special advantage for the application of this method of artificial neural network parameter optimization in engineering problems.
Кључне речи:
H∞ filter / Artificial neural network / Gaussian-type radial activation functions / Linearized Kalman filter / Parameter uncertainty / Measurement noises / Gaussian distributionИзвор:
Proceedings of the 6th International Working Conference ”Total Quality Management – Advanced and Intelligent Approaches”, 2011, 163-168Издавач:
- JUQS d.o.o. Beograd
Финансирање / пројекти:
Колекције
Институција/група
Mašinski fakultetTY - CONF AU - Vuković, Najdan AU - Miljković, Zoran AU - Babić, Bojan AU - Bojović, Božica PY - 2011 UR - https://machinery.mas.bg.ac.rs/handle/123456789/4478 AB - This paper analyzes the application of the H∞ filter for the optimization of the parameters of an artificial neural network with Gaussian-type radial activation functions. The analysis showed that the H∞ filter generates better estimates of the parameters of the artificial neural network than the linearized Kalman filter in problems where there is significant initial parameter uncertainty, insufficient knowledge of system/process characteristics, process noise, and measurement noise. Unlike the linearized Kalman filter, the H∞ filter does not rely on the assumption that process and measurement noises are subject to Gaussian distribution, which is a special advantage for the application of this method of artificial neural network parameter optimization in engineering problems. PB - JUQS d.o.o. Beograd C3 - Proceedings of the 6th International Working Conference ”Total Quality Management – Advanced and Intelligent Approaches” T1 - Training of Radial Basis Function Networks with H∞ Filter-Initial Simulation Results EP - 168 SP - 163 UR - https://hdl.handle.net/21.15107/rcub_machinery_4478 ER -
@conference{ author = "Vuković, Najdan and Miljković, Zoran and Babić, Bojan and Bojović, Božica", year = "2011", abstract = "This paper analyzes the application of the H∞ filter for the optimization of the parameters of an artificial neural network with Gaussian-type radial activation functions. The analysis showed that the H∞ filter generates better estimates of the parameters of the artificial neural network than the linearized Kalman filter in problems where there is significant initial parameter uncertainty, insufficient knowledge of system/process characteristics, process noise, and measurement noise. Unlike the linearized Kalman filter, the H∞ filter does not rely on the assumption that process and measurement noises are subject to Gaussian distribution, which is a special advantage for the application of this method of artificial neural network parameter optimization in engineering problems.", publisher = "JUQS d.o.o. Beograd", journal = "Proceedings of the 6th International Working Conference ”Total Quality Management – Advanced and Intelligent Approaches”", title = "Training of Radial Basis Function Networks with H∞ Filter-Initial Simulation Results", pages = "168-163", url = "https://hdl.handle.net/21.15107/rcub_machinery_4478" }
Vuković, N., Miljković, Z., Babić, B.,& Bojović, B.. (2011). Training of Radial Basis Function Networks with H∞ Filter-Initial Simulation Results. in Proceedings of the 6th International Working Conference ”Total Quality Management – Advanced and Intelligent Approaches” JUQS d.o.o. Beograd., 163-168. https://hdl.handle.net/21.15107/rcub_machinery_4478
Vuković N, Miljković Z, Babić B, Bojović B. Training of Radial Basis Function Networks with H∞ Filter-Initial Simulation Results. in Proceedings of the 6th International Working Conference ”Total Quality Management – Advanced and Intelligent Approaches”. 2011;:163-168. https://hdl.handle.net/21.15107/rcub_machinery_4478 .
Vuković, Najdan, Miljković, Zoran, Babić, Bojan, Bojović, Božica, "Training of Radial Basis Function Networks with H∞ Filter-Initial Simulation Results" in Proceedings of the 6th International Working Conference ”Total Quality Management – Advanced and Intelligent Approaches” (2011):163-168, https://hdl.handle.net/21.15107/rcub_machinery_4478 .