Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Springer eBooks

Error estimates on averages of correlated data

Henrik Flyvbjerg

Published Apr 6, 2008• 721 citations
Publisher
DOI
10.1007/bfb0105461

Related papers

Error estimates on averages of correlated data

Henrik Flyvbjerg, Henrik Gordon Petersen

We describe how the true statistical error on an average of correlated data can be obtained with ease and efficiency by a renormalization group method. The method is illustrated with numerical and analytical examples, having finite as well as infinite range correlations.

The Journal of Chemical Physics•Jul 1, 1989•1,567 citations

Error Bounds for Smooth Correlations

Emil V. Prodan

SpringerBriefs in mathematical physics•Jan 1, 2017•1 citation

Dynamic Correlations and the Response Function

WORLD SCIENTIFIC eBooks•Feb 1, 2008•0 citations

Self-consistent tomography and measurement-device independent cryptography

I. D. Moore, S. J. van Enk

A recurring problem in quantum mechanics is to estimate either the state of a quantum system or the measurement operator applied to it. If we wish to estimate both, then the difficulty is that the state and the measurement always appear together: to estimate the state, we must use a measurement; to estimate the measurement operator, we must use a state. The data of such quantum estimation experiments come in the form of measurement frequencies. Ideally, the measured average frequencies can be attributed to an average state and an average measurement operator. If this is not the case, we have correlated state-preparation-and-measurement (SPAM) errors. We extend some tests developed to detect such correlated errors to apply to a cryptographic scenario in which two parties trust their individual states but not the measurement performed on the joint state.

arXiv:2006.06559•International Journal of Quantum Information•Jul 4, 2020•3 citations

Optimal estimation of a physical observable’s expectation value for pure states

A. Hayashi, M. Horibe, T. Hashimoto

We study the optimal way to estimate the quantum expectation value of a physical observable when a finite number of copies of a quantum pure state are presented. The optimal estimation is determined by minimizing the squared error averaged over all pure states distributed in a unitary invariant way. We find that the optimal estimation is ``biased'' though the optimal measurement is given by successive projective measurements of the observable. The optimal estimate is not the sample average of observed data, but the arithmetic average of observed and ``default nonobserved'' data, with the latter consisting of all eigenvalues of the observable.

arXiv:quant-ph/0512037•Physical Review A•Jun 14, 2006•5 citations

Matrix elements between correlated Gaussians

CRC Press eBooks•Feb 6, 2003•0 citations
A recurring problem in quantum mechanics is to estimate either the state of a quantum system or the measurement operator applied to it. If we wish to estimate both, then the difficulty is that the state and the measurement always appear together: to estimate the state, we must use a measurement; to estimate the measurement operator, we must use a state. The data of such quantum estimation experiments come in the form of measurement frequencies. Ideally, the measured average frequencies can be attributed to an average state and an average measurement operator. If this is not the case, we have correlated state-preparation-and-measurement (SPAM) errors. We extend some tests developed to detect such correlated errors to apply to a cryptographic scenario in which two parties trust their individual states but not the measurement performed on the joint state.
We study the optimal way to estimate the quantum expectation value of a physical observable when a finite number of copies of a quantum pure state are presented. The optimal estimation is determined by minimizing the squared error averaged over all pure states distributed in a unitary invariant way. We find that the optimal estimation is ``biased'' though the optimal measurement is given by successive projective measurements of the observable. The optimal estimate is not the sample average of observed data, but the arithmetic average of observed and ``default nonobserved'' data, with the latter consisting of all eigenvalues of the observable.