A Sequence Well Dispersed in the Unit Square (Q3801667)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 4069015
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | A Sequence Well Dispersed in the Unit Square |
scientific article; zbMATH DE number 4069015 |
Statements
1988
0 references
dispersion
0 references
quasirandom search methods
0 references
low-dispersion sequence
0 references
dyadic sequence
0 references
A Sequence Well Dispersed in the Unit Square (English)
0 references
0.8428693413734436
0 references
0.8415759801864624
0 references
0.8278437256813049
0 references
0.8217155337333679
0 references
The dispersion is a measure for the irregularity of distribution of sequences which is important in quasirandom search methods for global optimization. The paper describes a recursive method of generating points in the unit square which form a low-dispersion sequence. All the points are dyadic, i.e. their coordinates are dyadic fractions, and so they are very convenient for machine implementation. The dispersion \(d_ N\) of the first N terms of this sequence satisfies \(d_ N=O(N^{-1/2}),\) which is known to be the lowest possible order of magnitude for the dispersion of any sequence in the unit square. The author's sequence has uniformly lower dispersion than any other known dyadic sequence, but it does not reach the overall best low-dispersion sequence constructed by the reviewer [Mathematical statistics and applications, Proc. 4th Pannonian Symp. Math. Stat., Bad Tatzmannsdorf/Austria 1983, Vol. B, 251- 267 (1985; Zbl 0603.65043)]. Misprints: in Figure 2b replace \(2^{k-1}\) by \(2^{k+1}\) and in reference [14] the correct name of the second author is R. B. Statnikov.
0 references