Tampere University of Technology

TUTCRIS Research Portal

On Renyi's entropy estimation with one-dimensional Gaussian kernels

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

On Renyi's entropy estimation with one-dimensional Gaussian kernels. / Sarbu, Septimia.

2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2016. p. 4408-4412.

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Sarbu, S 2016, On Renyi's entropy estimation with one-dimensional Gaussian kernels. in 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, pp. 4408-4412, IEEE International Conference on Acoustics, Speech and Signal Processing, 1/01/00. https://doi.org/10.1109/ICASSP.2016.7472510

APA

Sarbu, S. (2016). On Renyi's entropy estimation with one-dimensional Gaussian kernels. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 4408-4412). IEEE. https://doi.org/10.1109/ICASSP.2016.7472510

Vancouver

Sarbu S. On Renyi's entropy estimation with one-dimensional Gaussian kernels. In 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE. 2016. p. 4408-4412 https://doi.org/10.1109/ICASSP.2016.7472510

Author

Sarbu, Septimia. / On Renyi's entropy estimation with one-dimensional Gaussian kernels. 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 2016. pp. 4408-4412

Bibtex - Download

@inproceedings{67f3e499eb94429ea9c4cef0dc1ff841,
title = "On Renyi's entropy estimation with one-dimensional Gaussian kernels",
abstract = "R{\'e}nyi's entropies play a significant role in many signal processing applications. Plug-in kernel density estimation methods have been employed to estimate such entropies with good results. However, they become computationally intractable in higher dimensions, because of the requirement to store intermediate probability density values for a large number of data points. We propose a method to reduce the number of the samples in a plug-in kernel density estimation method for R{\'e}nyi's entropies of real exponents and to improve the result of the standard plug-in kernel density method. To this end, we derive a univariate estimator, using an Hermite expansion of sums of Gaussian kernels and a hierarchical clustering of the samples. On simulated data from a univariate Gaussian distribution, our method performs better than a k-nearest neigbour algorithm and other kernel density estimation methods.",
keywords = "Gaussian kernels, Hermite expansion, hierarchical clustering, R{\'e}nyi's entropy estimation",
author = "Septimia Sarbu",
year = "2016",
month = "5",
day = "18",
doi = "10.1109/ICASSP.2016.7472510",
language = "English",
isbn = "9781479999880",
publisher = "IEEE",
pages = "4408--4412",
booktitle = "2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - On Renyi's entropy estimation with one-dimensional Gaussian kernels

AU - Sarbu, Septimia

PY - 2016/5/18

Y1 - 2016/5/18

N2 - Rényi's entropies play a significant role in many signal processing applications. Plug-in kernel density estimation methods have been employed to estimate such entropies with good results. However, they become computationally intractable in higher dimensions, because of the requirement to store intermediate probability density values for a large number of data points. We propose a method to reduce the number of the samples in a plug-in kernel density estimation method for Rényi's entropies of real exponents and to improve the result of the standard plug-in kernel density method. To this end, we derive a univariate estimator, using an Hermite expansion of sums of Gaussian kernels and a hierarchical clustering of the samples. On simulated data from a univariate Gaussian distribution, our method performs better than a k-nearest neigbour algorithm and other kernel density estimation methods.

AB - Rényi's entropies play a significant role in many signal processing applications. Plug-in kernel density estimation methods have been employed to estimate such entropies with good results. However, they become computationally intractable in higher dimensions, because of the requirement to store intermediate probability density values for a large number of data points. We propose a method to reduce the number of the samples in a plug-in kernel density estimation method for Rényi's entropies of real exponents and to improve the result of the standard plug-in kernel density method. To this end, we derive a univariate estimator, using an Hermite expansion of sums of Gaussian kernels and a hierarchical clustering of the samples. On simulated data from a univariate Gaussian distribution, our method performs better than a k-nearest neigbour algorithm and other kernel density estimation methods.

KW - Gaussian kernels

KW - Hermite expansion

KW - hierarchical clustering

KW - Rényi's entropy estimation

U2 - 10.1109/ICASSP.2016.7472510

DO - 10.1109/ICASSP.2016.7472510

M3 - Conference contribution

SN - 9781479999880

SP - 4408

EP - 4412

BT - 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

PB - IEEE

ER -