Salvato in:
Dettagli Bibliografici
Autore principale: Ellerman, David
Natura: Preprint
Pubblicazione: 2014
Soggetti:
Accesso online:https://arxiv.org/abs/1407.4345
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910591394775040
author Ellerman, David
author_facet Ellerman, David
contents By using a new way to encode Boolean functions in a reversible gate, an algorithm is developed in quantum computing over Z_2, symbolized QC/2, (as opposed to QC over C) that needs only one function evaluation to solve the Grover Database Search Problem of finding a designated record among 2^m records for any m. In the usual Grover algorithm in quantum computing over C, one needs essentially Sqrt(2^m) function evaluations as opposed to the average of (2^m)/2 functions evaluations needed in the classical algorithm. The one function evaluation of the QC/2 algorithm (for any m) represents such a super speedup, even over the Grover algorithm in QC/C, that one feels something has gone awry. Indeed, our analysis of the transparent calculations of Boolean functions over Z_2 shows that the classical algorithm is just repackaged in a rather obvious way in the single function evaluation of the QC/2 algorithm--whereas the calculations are hidden and non-transparent in the Grover QC/C algorithm using C. The conclusion in both cases (which is rather obvious in the QC/2 case) is that "counting function evaluations" is a false coin to measure speedup in the comparison between quantum and classical computing.
format Preprint
id arxiv_https___arxiv_org_abs_1407_4345
institution arXiv
publishDate 2014
record_format arxiv
spellingShingle In Quantum Computing Speedup Illusory?: The False Coin of "Counting Function Evaluations"
Ellerman, David
Quantum Physics
81P68
By using a new way to encode Boolean functions in a reversible gate, an algorithm is developed in quantum computing over Z_2, symbolized QC/2, (as opposed to QC over C) that needs only one function evaluation to solve the Grover Database Search Problem of finding a designated record among 2^m records for any m. In the usual Grover algorithm in quantum computing over C, one needs essentially Sqrt(2^m) function evaluations as opposed to the average of (2^m)/2 functions evaluations needed in the classical algorithm. The one function evaluation of the QC/2 algorithm (for any m) represents such a super speedup, even over the Grover algorithm in QC/C, that one feels something has gone awry. Indeed, our analysis of the transparent calculations of Boolean functions over Z_2 shows that the classical algorithm is just repackaged in a rather obvious way in the single function evaluation of the QC/2 algorithm--whereas the calculations are hidden and non-transparent in the Grover QC/C algorithm using C. The conclusion in both cases (which is rather obvious in the QC/2 case) is that "counting function evaluations" is a false coin to measure speedup in the comparison between quantum and classical computing.
title In Quantum Computing Speedup Illusory?: The False Coin of "Counting Function Evaluations"
topic Quantum Physics
81P68
url https://arxiv.org/abs/1407.4345