ISSN 2071-8594

Russian academy of sciences

Editor-in-Chief

Gennady Osipov

A.A. Bedrintsev, V.V. Chepyzhov, S.S. Chernova. Extreme ellipsoids as approximations of design space in data predictive metamodeling problems

Abstract.

This paper proposes an approach to obtaining of the set of admissible values of the optimization variables (design space) in the form of extreme ellipsoids describing a given set of points and inscribed in a given set of linear constraints. Considered ellipsoids include Principal Component’s ellipsoid, minimal volume ellipsoid and ellipsoid with minimal trace of its matrix containing given points. We have developed the procedures which change ellipsoid built based on points set exclusively in order to inscribe it into polyhedron. Ellipsoids are constructed by solving corresponding optimization problems which are formulated as convex programming problems using linear matrix inequalities.

Keywords:

data representation; extreme ellipsoids; convex optimization; linear matrix inequalities; Principal Components Analysis.

PP. 35-44.

REFERENCES

1. Kuleshov A.P. Tekhnologiya bystrogo vychisleniya kharakteristik slozhnykh tekhnicheskikh obektov/ Informatsionnye tekhnologii, pril. № 3, M., 2006, s. 4 - 11.
2. Kuleshov A.P. Kognitivnye tekhnologii v adaptivnykh modelyakh slozhnykh obektov/ Informatsionnye tekhnologii i vychislitelnye sistemy, v. 1, 2008, s. 18 – 29.
3. Kuleshov A.P. Metamodeli v kompyuternykh sistemakh proektirovaniya. Trudy Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii «Informatsionnye tekhnologii i matematicheskoe modelirovanie sistem» (ITMM, 19 – 26 sentyabrya 2010 g., Frantsiya). M.: Uchrezhdenie rossiyskoy akademii nauk Tsentr informatsionnykh tekhnologiy v proektirovanii RAN, 2010. s. 209-210.
4. Forrester, A.I.J., Sobester A., Keane A.J. Engineering Design via Surrogate Modelling. A Practical Guide. Wiley, New-York, 2008.
5. Kuleshov A.P., Bernstein A.V. Cognitive technologies in adaptive models of complex plants. Information Control Problems in Manufacturing. 2009, Vol. 13, Part 1, pp. 1441 – 1452.
6. Wang G., Shan S. Review of Metamodeling Techniques in Support of Engineering Design Optimization. J. Mech. Des. 2007, 129(3), pp. 370-381.
7. Bernshteyn A.V., Kuleshov A.P. Matematicheskie metody postroeniya metamodeley. Trudy Tretey mezhdunarodnoy konferentsiya «Sistemnyy analiz i informatsionnye tekhnologii» (SAIT-2009, 14-18 sentyabrya 2009 g., Zvenigorod, Rossiya). 2009, s. 756 – 768.
8. Kuleshov A.P., Bernshteyn A.V. Problemy intellektualnogo analiza dannykh dlya sozdaniya metamodeley. Trudy Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii «Informatsionnye tekhnologii i matematicheskoe modelirovanie sistem» (ITMM, 19 – 26 sentyabrya 2010 g., Frantsiya). M.: Tsentr informatsionnykh tekhnologiy v proektirovanii RAN, 2010. c. 222 - 223.
9. Bernshteyn A.V., Burnaev Ye.V., Kuleshov A.P. Intellektualnyy analiz dannykh v metamodelirovanii. Neyroinformatika, ee prilozheniya i analiz dannykh. Materialy XVII Vserossiyskogo seminara (2 – 4 oktyabrya 2009 g.), Krasnoyarsk, IPK SFU, 2009, s. 23 – 28.
10. Kuleshov A.P., Bernshteyn A.V. Predskazatelnoe metamodelirovanie i intellektualnyy analiz dannykh. Dvenadtsataya natsionalnaya konferentsiya po iskusstvennomu intellektu s mezhdunarodnym uchastiem (KII-2010). Trudy konferentsii. - T. 1. - M.: Fizmatlit, 2010.
11. Bernshteyn A.V., Burnaev Ye.V., Kuleshov A.P. Ob odnoy metodologii postroeniya approksimatsiy mnogomernykh zavisimostey // Trudy «Plenarnye i izbrannye doklady» IV mezhdunarodnoy konferentsii «Parallelnye vychisleniya i zadachi upravleniya», Moskva, C. 56 – 62, 2008.
12. Bernshteyn A.V., Kuleshov A.P. Kognitivnye tekhnologii v probleme snizheniya razmernosti opisaniya geometricheskikh obektov // Informatsionnye tekhnologii i vychislitelnye sistemy. 2008, №2.
13. Bernshteyn A.V., Burnaev Ye.V. Zadacha snizheniya razmernosti v predskazatelnom modelirovanii. Sbornik dokladov 9-y mezhdunarodnoy konferentsii
«Intellektualizatsiya obrabotki informatsii» (IOI-9), Chernogoriya, g. Budva, 2012. M.: Torus Press, 2012.
14. Bernshteyn A.V. Snizhenie razmernosti v predskazatelnom modelirovanii: novye zadachi i metody. Sbornik trudov konferentsii "Informatsionnye tekhnologii i sistemy - 2012". 20-24 avgusta 2012, g. Rossiya, g. Petrozavodsk. M.: IPPI RAN, 2012.
15. Bernstein A.V., Kuleshov A.P. Low-Dimensional Data Representation in Data Analysis, Lecture Notes in Artificial Intelligence, Springer International Publishing, Switzerland, vol. 8774, p. 47-58, 2014.
16. S.S. Chernova, Ye.P. Ivanova. Snizhenie razmernosti slozhnykh geometricheskikh obektov pri nalichii chastnykh parametricheskikh modeley. Iskusstvennyy intellekt i prinyatie resheniy, 2009, №3, s. 53 – 58.
17. Bernstein A.V., Burnaev E.V., Chernova S.S., F. Zhu, N. Qin. Comparison of Three Geometric Parameterization methods and Their Effect on Aerodynamic Optimization. Proceedings of International Conference on Evolutionary and Deterministic Methods for Design, Optimization and Control with Applications to Industrial and Societal Problems (Eurogen 2011). Capua, Italy, September 14 –
16, 2011. http://www.eurogen2011.cira.it/
18. Bernstein, A., Kuleshov, A., Sviridenko, Yu., Vyshinsky, V. Fast Aerodynamic Model for Design Technology. In: Proceedings of West-East High Speed Flow Field Conference. Nov. 19-22, 2007, Moscow, Russia, http://wehsff.imamod.ru/pages/s7.htm.
19. A.V. Bernshteyn, V.V. Vyshinskiy, A.P. Kuleshov, Yu.N. Sviridenko. Bystryy metod aerodinamicheskogo rascheta dlya zadach proektirovaniya // Trudy Tsentralnogo aerodinamicheskogo instituta im. prof. N.Ye. Zhukovskogo. 2008. Vypusk №2678 «Primenenie iskusstvennykh neyronnykh setey v zadachakh prikladnoy aerodinamiki». M.: TsAGI.
20. Jollie T. Principal Component Analysis. New-York, Springer, 2002.
21. Ye.V. Burnaev, S.S. Chernova. Ob iterativnom algoritme podscheta glavnykh komponent // Informatsionnye protsessy. 2008. Tom 8, №2, s. 99-107.
22. Boyd S. Convex Optimization. Cambridge: University Press, 2004.
23. Yu.Ye. Nesterov. Metody vypukloy optimizatsii. M.: MTsMNO, 2010.
24. C. Bradford Barber et al. The Quickhull Algorithm for Convex Hull // ACM Transactions on Mathematical Software. 1996. Vol. 22, No. 4, pp. 469-483.
25. Wei-Cheng Chang, Ching-Pei Lee and Chih-Jen Lin, A Revisit to Support Vector Data Description. Technical report, 2013.
26. Bo Liu et al.. SVDD-based outlier detection on uncertain data. Knowledge and Information Systems, 2013. Volume 34, Issue 3, pp. 597-618.
27. Qing Ai, Ji Zhao, Yuping Qin. Study of Small Sphere and Large Margin Hypersphere Support Vector Machine. Journal of Convergence Information Technology(JCIT) . Volume8, Number7, April. 2013. doi:10.4156/jcit.vol8.issue7.143
28. Y. Zhang et al., Distributed online outlier detection in wireless sensor networks using ellipsoidal support vector machine, Ad Hoc Netw. (2012), http://dx.doi.org/10.1016/j.adhoc.2012.11.001.
29. Bedrintsev A.A., Chepyzhov V.V. Predstavlenie dannykh s pomoshchyu ekstremalnykh ellipsoidov // Materialy konferentsii ITiS, Kaliningrad, 2013.
30. Stephen Boyd et al. Linear Matrix Inequalities in System and Control Theory // SIAM studies in applied mathematics. 1994. Vol. 15.
31. CVX: Matlab Software for Disciplined Convex Programming, http://cvxr.com/cvx.