Specific process models derived from extremely small data sets and general process models
Abstract
Definition of a needed particular process model is based on a combination of weighted known general process models and standard error minimization. The known general process models correspond to the biological processes of growing. The standard error is computed using new data and an ensemble of generated models. General models are based on polynomial functions and neural networks. Applications of polynomial functions of second, third and fourth degrees is analyzed. Supervised learning of the neural networks is based on the Levenberg-Marquardt algorithm. A very brief comment on the Vapnik-Chervonenkis dimension as an important parameter in modern learning theory, is also done in view of the analyzed cases.
Keywords:
Neural networks / Process modeling / Regession problem / Vapnik-Chervonenkis dimensionSource:
2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL, 2004, 267-272Publisher:
- 2004 Seventh Seminar on Neural Network Applications in Elecrtical Engineering, NEUREL 2004
Collections
Institution/Community
Poljoprivredni fakultetTY - CONF AU - Radonja, P. AU - Stanković, S. AU - Popović, Zoran PY - 2004 UR - http://aspace.agrif.bg.ac.rs/handle/123456789/685 AB - Definition of a needed particular process model is based on a combination of weighted known general process models and standard error minimization. The known general process models correspond to the biological processes of growing. The standard error is computed using new data and an ensemble of generated models. General models are based on polynomial functions and neural networks. Applications of polynomial functions of second, third and fourth degrees is analyzed. Supervised learning of the neural networks is based on the Levenberg-Marquardt algorithm. A very brief comment on the Vapnik-Chervonenkis dimension as an important parameter in modern learning theory, is also done in view of the analyzed cases. PB - 2004 Seventh Seminar on Neural Network Applications in Elecrtical Engineering, NEUREL 2004 C3 - 2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL T1 - Specific process models derived from extremely small data sets and general process models EP - 272 SP - 267 DO - 10.1109/neurel.2004.1416592 ER -
@conference{ author = "Radonja, P. and Stanković, S. and Popović, Zoran", year = "2004", abstract = "Definition of a needed particular process model is based on a combination of weighted known general process models and standard error minimization. The known general process models correspond to the biological processes of growing. The standard error is computed using new data and an ensemble of generated models. General models are based on polynomial functions and neural networks. Applications of polynomial functions of second, third and fourth degrees is analyzed. Supervised learning of the neural networks is based on the Levenberg-Marquardt algorithm. A very brief comment on the Vapnik-Chervonenkis dimension as an important parameter in modern learning theory, is also done in view of the analyzed cases.", publisher = "2004 Seventh Seminar on Neural Network Applications in Elecrtical Engineering, NEUREL 2004", journal = "2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL", title = "Specific process models derived from extremely small data sets and general process models", pages = "272-267", doi = "10.1109/neurel.2004.1416592" }
Radonja, P., Stanković, S.,& Popović, Z.. (2004). Specific process models derived from extremely small data sets and general process models. in 2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL 2004 Seventh Seminar on Neural Network Applications in Elecrtical Engineering, NEUREL 2004., 267-272. https://doi.org/10.1109/neurel.2004.1416592
Radonja P, Stanković S, Popović Z. Specific process models derived from extremely small data sets and general process models. in 2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL. 2004;:267-272. doi:10.1109/neurel.2004.1416592 .
Radonja, P., Stanković, S., Popović, Zoran, "Specific process models derived from extremely small data sets and general process models" in 2004 Seventh Seminar on Neural Network Applications in Electrical Engineering - Proceedings, NEUREL (2004):267-272, https://doi.org/10.1109/neurel.2004.1416592 . .