Please use this identifier to cite or link to this item: https://doi.org/10.2316/Journal.201.2014.2.201-2567
Title: Model assessment through renormalization group in statistical learning
Authors: Wang, Q.-G. 
Yu, C.
Zhang, Y.
Keywords: Binary classification
Model assessment
Renor-malization Group
Statistical learning
Support vector machines
Issue Date: 14-Apr-2014
Citation: Wang, Q.-G., Yu, C., Zhang, Y. (2014-04-14). Model assessment through renormalization group in statistical learning. Control and Intelligent Systems 42 (2) : 126-135. ScholarBank@NUS Repository. https://doi.org/10.2316/Journal.201.2014.2.201-2567
Abstract: This paper proposes a new method for model assessment based on Renormalization Group (RG). RG is applied to the original data set to obtain the transformed data set with the majority rule to set its labels. The assessment is first performed on the data level without invoking any learning method, and the consistency and non-randomness indices are defined by comparing two data sets to reveal informative content of the data. When the indices indicate informative data, the next assessment is carried out at the model level, and the predictions are compared between two models learnt from the original and transformed data sets, respectively. The model consistency and reliability indices are introduced accordingly. Unlike cross-validation and other standard methods in the literature, the proposed method creates a new data set and data assessment. Besides, it requires only two models and thus less computational burden for model assessment. The proposed method is illustrated with academic and practical examples.
Source Title: Control and Intelligent Systems
URI: http://scholarbank.nus.edu.sg/handle/10635/82708
ISSN: 14801752
DOI: 10.2316/Journal.201.2014.2.201-2567
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Page view(s)

31
checked on Oct 27, 2018

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.