论文信息 - A note on pseudorandom number generators

A note on pseudorandom number generators

Abstract PRNGs may be compared to antibiotics. Every type of generator has its unwanted side-effects. There are no safe generators. Good random number generators are characterized by theoretical support, convincing empirical evidence, and positive pratical aspects. They will produce correct results in many - though not all - simulations. Important open questions in this field concern reliable parallelization, the creation of good generators on demand, and the mathematical foundation of forecasting the empirical performance by theoretical figures of merit. Three safety-measures for numerical practice are recommended: (a) check simulation results with widely different generators before taking them seriously, (b) avoid to combine, vectorize, or parallelize PRNGs without theoretical and empirical support, and (c) get to know the properties of your preferred PRNGs.

Peter Hellekalek | P. Hellekalek

[1] T. Ala‐Nissila,et al. Physical models as tests of randomness. , 1995, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[2] Makoto Matsumoto,et al. Twisted GFSR generators , 1992, TOMC.

[3] Donald Ervin Knuth,et al. The Art of Computer Programming , 1968 .

[4] P. Hellekalek. Good random number generators are (not so) easy to find , 1998 .

[5] Pierre L'Ecuyer,et al. Maximally equidistributed combined Tausworthe generators , 1996, Math. Comput..

[6] Harald Niederreiter,et al. New Developments in Uniform Pseudorandom Number and Vector Generation , 1995 .

[7] Makoto Matsumoto,et al. Twisted GFSR generators II , 1994, TOMC.

[8] Pierre L'Ecuyer,et al. Testing random number generators , 1992, WSC '92.

[9] L. Devroye. Non-Uniform Random Variate Generation , 1986 .

[10] Pierre L'Ecuyer,et al. Random Number Generators and Empirical Tests , 1998 .

[11] Peter Hellekalek. Inversive pseudorandom number generators: concepts, results and links , 1995, WSC '95.

[12] Pierre L'Ecuyer,et al. Combined Multiple Recursive Random Number Generators , 1995, Oper. Res..

[13] G. S. Fishman. Multiplicative congruential random number generators with modulus 2^{}: an exhaustive analysis for =32 and a partial analysis for =48 , 1990 .