Precisely evaluating of discrepancy

Piotr Karnasiewicz

Abstract


Discrepancy is a measure of quality of samples distribution expressed as a supremum. In general, estimating this supremum relies on comparing volumes of many randomly chosen sets from some given family to quantity of samples belonging to these sets. In this paper, a numerical method of precise determinating discrepancy using the most popular families of sets will be introduced.

Keywords


discrepancy; sampling

Full Text:

PDF (Polski)

References


Foley J. D., van Dam A., Feiner S. K., Hughes J. F., Philips R. L.: Wprowadzenie do grafiki komputerowej. WNT, Warszawa 2001.

Gentle J. E.: Random Number Generation and Monte Carlo Methods. Springer, Berlin 1998.

Niederreiter H.: Random Number Generation and Quasi-Monte Carlo Methods. Society for Industrial and Applied Mathematics, 1992.

Pharr M., Humphreys G.: Physically Based Rendering. From theory to implementation. Elsevier, 2003.

Tadeusiewicz R., Korohoda P.: Komputerowa analiza i przetwarzanie obrazów. Wydawnictwo Fundacji Postępu Komunikacji, 1997.

Veach E.: Robust Monte Carlo methods for light transport simulation. Rozprawa doktorska, Stanford University, 1997.




DOI: http://dx.doi.org/10.21936/si2008_v29.n4A.518