Abstract
The aim of this paper is to highlight the potential of Machine Learning for parameter screening in computer simulations, presenting alternative approaches to automatic parameter ranking and screening. This is indeed a fundamental step in the development of a simulator, because it allows reducing the dimensionality of the parameter set, making model tuning more efficient. With parameter ranking we denote the process of measuring the relevance of the parameters for accurately simulating a phenomenon, while with parameter screening we denote the choice of a specific subset of parameters to be used for model tuning. We will present ranking techniques based on Logistic Regression and Multilayer Perceptron, and a simple procedure for going from ranking to screening. Our techniques have been validated against a helicopter simulator case-study but the techniques do not rely on any domain-specific feature or assumption.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Morgan, P.J., Cleave-Hogg, D., Desousa, S.: Applying theory to practice in undergraduate education using high fidelity simulation. Med. Teach. 28(1), e10–e15 (2006)
Lewis, J.H., Jiang, S.B.: A theoretical model for respiratory motion artifacts in free-breathing CT scans. Phys. Med. Biol. 54(3), 745–755 (2009)
Zhang, A.: One-factor-at-a-time Screening Designs for Computer Experiments, SAE Technical Paper 2007-01-1660 (2007), doi:10.4271/2007-01-1660
Bettonvil, B., Kleijnen, J.P.C.: Searching for important factors in simulation models with many factors: Sequential bifurcation. European Journal of Operational Research 96(1), 180–194 (1997)
Last, M., Luta, G., Orso, A., Porter, A., Young, S.: Pooled ANOVA. Computational Statistics & Data Analysis 52(12), 5215–5228 (2008)
Fisher, R.A.: The design of experiments, xi 251 p. Oliver & Boyd, Oxford (1935)
Harrel, F.: Regression Modeling Strategies. Springer (2001)
Bishop, C.: Pattern Recognition and Machine Learning, pp. 217–218. Springer Science+Business Media, LLC (2006)
Nelder, J., Wedderburn, R.: Generalized Linear Models. Journal of the Royal Statistical Society. Series A (General) 135(3), 370–384 (1972)
Le Cessie, S., Van Houwelingen, J.C.: Ridge estimators in Logistic Regression. Applied Statistics (1992)
Haykin, S.: Neural Networks: A Comprehensive Foundation, 2nd edn. Prentice Hall (1998) ISBN 0-13273350-1
Hall, M., Eibe, F., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.: The WEKA Data Mining Software: An Update. SIGKDDExplorations 11(1) (2009)
Bergmeir, C., Benìtez, J.M.: Neural Networks in R Using the Stuttgart Neural Network Simulator: RSNNS. Journal of Statistical Software 46(7) (2012)
Rosenblatt, F.: The perceptron: a probabilistic model for information storage and organization in the brain. Psychological Review 65, 386–408 (1958)
Cybenko, G.: Approximations by superpositions of sigmoidal functions. Mathematics of Control, Signals, and Systems 2(4), 303–314 (1989)
Hinton, G.E., Osindero, S.: Yee-Whye The: A fast learning algorithm for deep belief nets. Neural Computation 18(7), 1527–1554 (2006)
Rumelhart, D.E., Hinton, G.E., Williams, R.J.: Learning representations by back-propagatingerrors. Nature 323(6088), 533–536 (1986), doi:10.1038/323533a0
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Hessel, M., Ortalli, F., Borgatelli, F. (2014). Machine Learning for Parameter Screening in Computer Simulations. In: Hodicky, J. (eds) Modelling and Simulation for Autonomous Systems. MESAS 2014. Lecture Notes in Computer Science, vol 8906. Springer, Cham. https://doi.org/10.1007/978-3-319-13823-7_27
Download citation
DOI: https://doi.org/10.1007/978-3-319-13823-7_27
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-13822-0
Online ISBN: 978-3-319-13823-7
eBook Packages: Computer ScienceComputer Science (R0)