Breakthrough in Interval Data Fitting I. The Role of Hausdorff Distance

• Autorzy: Gutowski M.W.
• Konferencja: Evolutionary Computation and Global Optimization 2009 / National Conference (12 ; 1-3.06.2009 ; Zawoja, Poland)
• Języki publikacji: EN

Abstrakty

EN

This is the first of two papers describing the process of fitting experimental data under interval uncertainty. Probably the most often encountered application of global optimization methods is finding the so called best fitted values of various parameters, as well as their uncertainties, based on experimental data. Here I present the methodology, designed from the very beginning as an interval-oriented tool, meant to replace to the large extent the famous Least Squares (LSQ) and other slightly less popular methods. Contrary to its classical counterparts, the presented method does not require any poorly justified prior assumptions, like smallness of experimental uncertainties or their normal (Gaussian) distribution. Using interval approach, we are able to fit rigorously and reliably not only the simple functional dependencies, with no extra effort when both variables are uncertain, but also the cases when the constitutive equation exists in implicit rather than explicit functional form. The magic word and a key to success of interval approach appears the Hausdorff distance.

Słowa kluczowe

EN

global optimization; interval data; interval computations; least squares method; Hausdorff distance; Hausdorff space

PL

optymalizacja globalna; dane interwałowe; przedział obliczeń; metoda najmniejszych kwadratów; odległość Hausdorffa; przestrzeń Hausdorffa

• Wydawca: Oficyna Wydawnicza Politechniki Warszawskiej
• Czasopismo: Prace Naukowe Politechniki Warszawskiej. Elektronika
• Rocznik: 2009
• Tom: z. 169
• Strony: 63--72
• Opis fizyczny: Bibliogr. 25 poz.

Twórcy

autor

Gutowski M.W.

• Institute of Physics, Polish Academy of Sciences, 02-668 Warszawa, Poland,

Bibliografia

• [1] The website http://cs.utep.edu/interval-comp/main.html is an excellent and up-to-date entry point to the wonderful world of interval computations.
• [2] Luc Jaulin and Eric Walter, Guaranteed Nonlinear Parameter Estimation via Interval Computations, Conf. on Numerical Analysis with Automatic Result Verification, Lafayette, Feb. 25th-March 3rd, 1993, pp. 61-75
• [3] Luc Jaulin and Eric Walter, Guaranteed nonlinear parameter estimation from bounded-error data via interval analysis, Math. and Comput. in Simulation 35, 123-137, 1993
• [4] Luc Jaulin and Eric Walter, Set Inversion via Interval Analysis for Nonlinear Bounded-error Estimation, Automatica 29, 1053, 1993
• [5] L. Jaulin and E. Walter, Guaranteed Parameter Bounding for Nonlinear Models with Uncertain Experimental Factors, Automatica 35, 849-856, 1999
• [6] Luc Jaulin, Interval constraint propagation with application to bounded-error estimation, Automatica 36, 1547, 2000
• [7] Humberto Muñoz and R.B. Kearfott, Interval Robustness in Nonsmooth Nonlinear Parameter Estimation, unpublished preprint, http://interval.louisiana.edu/preprints/2001_robustness.pdf
• [8] L. Jaulin and E. Walter, Nonlinear Bounded-Error Parameter Estimation Using Interval Computation, in: Granular computing: an emerging paradigm, Physica-Verlag GmbH Heidelberg, pp. 58-71, 2001
• [9] Marek W. Gutowski, Prosta dostatecznie gruba, Postępy Fizyki, 53(4), 181-192, 2002 (in Polish) (Fat enough straight line, Advances in Physics, bimonthly of Polish Physical Society)
• [10] Voschinin Alexander, Tyurin Alexander, Interval identification of time series parameters using readings with bounded errors, 5th International Scientific-Technical Conf. ProcessControl, Pordubice 2002, paper R-210, 7 pages
• [11] M.H. van Emden, Using the duality principle to improve lower bounds for the global minimum in nonconvex optimization, Second COCOS workshop on intervals and optimization, 2003 (published version: Using Propagation for Solving Complex Arithmetic Constraints, http://arxiv.org/abs/cs/0309018)
• [12] I. Braems, N. Ramdani, A. Boudenne, L. Jaulin, L. Ibos, E. Walter, and Y. Candau, New set-membership techniques for parameter estimation in presence of model uncertainty, Proc. of the 5th Int. Conf. on Inverse Problems in Engineering: Theory and Practice, Cambridge, UK, 11-15 July 2005
• [13] Sergei I. Zhilin, On Fitting Empirical Data under Interval Error, Reliable Computing 11, 433-442, 2005
• [14] Maarten van Emden, Constraint-Driven Global Optimization. 13th International Symposium on Scientific Computing Computer Arithmetic and Verified Numerical Computations SCAN'2008, El Paso, Texas, September 29-October 3, pp. 144-145, 2008
• [15] L. Jaulin, J-L. Godet, E. Walter, A. Elliasmine, and Y. Le Duff, Light scattering data analysis via set inversion, J. Phys. A: Math. Gen. 30, 7733-7738, 1993
• [16] Maëlenn Aufray, Adrien Brochier, and Wulff Possart, Set Inversion via Interval Analysis applied to dielectric spectroscopy, the talk given at SWIM 2008, June 19-20th, Montpelier, France
• [17] T. Barsan, D. Tiba, One hundred years since the introduction of the set distance by Dimitrie Pompeiu, in: System modeling and optimization. Springer, New York, pp. 35-39, 2006
• [18] Svetoslav M. Markov, Least-square approximations under interval input data, Contributions to Computer Arithmetic and Self-Validating Numerical Methods, C. Ulrich (editor), J.C. Balzer AG, Scientific Publishing Co. (C) IMACS 1990, pp. 133-147
• [19] G. William Walster and Vladik Kreinovich, For Uknown-but-Bounded Errors, Interval Estimates are Often Better Than Averaging, ACM SIGNUM Newsletter 31, 6-19, 1996, http://www.cs.utep.edu/vladik/1993/tr93-31b.ps.gz
• [20] Olga Kosheleva and Vladik Kreinovich, Error Estimation for Indirect Measurements: Interval Computation Problem Is (Slightly) Harder Than a Similar Probabilistic Computational Problem, Reliable Computing 5, 81-95, 1999
• [21] Jie Yang and R. Baker Kearfott Interval Linear and Nonlinear Regression - New Paradigms, Implementations, and Experiments or New Ways of Thinking About Data Fitting, talk given at the Seventh SIAM Conference on Optimization, May 20-22, 2002, Toronto, Canada, http://interval.louisiana.edu/preprints/2002_ SIAM_minisymposium.ps
• [22] R.E. Moore Interval Analysis Prentice Hall, Englewood Cliffs, NJ, 1966
• [23] Francesco Palumbo and Antonio Irpino, Multidimensional Interval-Data: Metrics and Factorial Analysis, Applied Stochastic Models and Data Analysis, Brest, France May 17-20, 2005, pp. 689-698, http://webhouse.unimc.it/economia/repo/39/ 689.pdf
• [24] Antonio Irpino and Rosanna Verde, Dynamic clustering of interval data using a Wasserstein-based distance, Pattern Recognition Letters 29, 1648-1658, 2008
• [25] Sergei P. Shary, A Surprising Approach in Interval Global Optimization, Reliable Computing 7, 497-505, 2001