Skip to main content

Correntropy–based density–preserving data sampling as an alternative to standard cross–validation.

Budka, M. and Gabrys, B., 2010. Correntropy–based density–preserving data sampling as an alternative to standard cross–validation. In: ICJNN 2010: International Joint Conference on Neural Networks. IEEE, 1-8.

This is the latest version of this eprint.

Full text available as:

[img]
Preview
PDF
PID1190807.pdf - Accepted Version

277kB

Official URL: http://ieeexplore.ieee.org/search/srchabstract.jsp...

DOI: 10.1109/IJCNN.2010.5596717

Abstract

Estimation of the generalization ability of a predictive model is an important issue, as it indicates expected performance on previously unseen data and is also used for model selection. Currently used generalization error estimation procedures like cross–validation (CV) or bootstrap are stochastic and thus require multiple repetitions in order to produce reliable results, which can be computationally expensive if not prohibitive. The correntropy–based Density Preserving Sampling procedure (DPS) proposed in this paper eliminates the need for repeating the error estimation procedure by dividing the available data into subsets, which are guaranteed to be representative of the input dataset. This allows to produce low variance error estimates with accuracy comparable to 10 times repeated cross–validation at a fraction of computations required by CV, which has been investigated using a set of publicly available benchmark datasets and standard classifiers.

Item Type:Book Section
ISBN:978-1-4244-6916-1
ISSN:1098-7576
Additional Information:ICJNN 2010: Barcelona Spain 18-23 July 2010
Group:Faculty of Science & Technology
ID Code:16572
Deposited By: Dr Marcin Budka
Deposited On:21 Oct 2010 18:13
Last Modified:14 Mar 2022 13:36

Available Versions of this Item

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -