Introduction to Extreme Seeking Entropy
Recently, the concept of evaluating an unusually large learning effort of an adaptive system to detect novelties in the observed data was introduced. The present paper introduces a new measure of the learning effort of an adaptive system. The proposed method also uses adaptable parameters. Instead o...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-01-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/22/1/93 |
| id |
doaj-b9ec1bda4660418b93ce01024d388dc7 |
|---|---|
| record_format |
Article |
| spelling |
doaj-b9ec1bda4660418b93ce01024d388dc72020-11-25T01:10:11ZengMDPI AGEntropy1099-43002020-01-012219310.3390/e22010093e22010093Introduction to Extreme Seeking EntropyJan Vrba0Jan Mareš1Department of Computing and Control Engineering, Faculty of Chemical Engineering, University of Chemistry and Technology, 166 28 Prague, Czech RepublicDepartment of Computing and Control Engineering, Faculty of Chemical Engineering, University of Chemistry and Technology, 166 28 Prague, Czech RepublicRecently, the concept of evaluating an unusually large learning effort of an adaptive system to detect novelties in the observed data was introduced. The present paper introduces a new measure of the learning effort of an adaptive system. The proposed method also uses adaptable parameters. Instead of a multi-scale enhanced approach, the generalized Pareto distribution is employed to estimate the probability of unusual updates, as well as for detecting novelties. This measure was successfully tested in various scenarios with (i) synthetic data, (ii) real time series datasets, and multiple adaptive filters and learning algorithms. The results of these experiments are presented.https://www.mdpi.com/1099-4300/22/1/93novelty detectionlearning systemlearningtime serieslearning entropyextreme seeking entropy |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Jan Vrba Jan Mareš |
| spellingShingle |
Jan Vrba Jan Mareš Introduction to Extreme Seeking Entropy Entropy novelty detection learning system learning time series learning entropy extreme seeking entropy |
| author_facet |
Jan Vrba Jan Mareš |
| author_sort |
Jan Vrba |
| title |
Introduction to Extreme Seeking Entropy |
| title_short |
Introduction to Extreme Seeking Entropy |
| title_full |
Introduction to Extreme Seeking Entropy |
| title_fullStr |
Introduction to Extreme Seeking Entropy |
| title_full_unstemmed |
Introduction to Extreme Seeking Entropy |
| title_sort |
introduction to extreme seeking entropy |
| publisher |
MDPI AG |
| series |
Entropy |
| issn |
1099-4300 |
| publishDate |
2020-01-01 |
| description |
Recently, the concept of evaluating an unusually large learning effort of an adaptive system to detect novelties in the observed data was introduced. The present paper introduces a new measure of the learning effort of an adaptive system. The proposed method also uses adaptable parameters. Instead of a multi-scale enhanced approach, the generalized Pareto distribution is employed to estimate the probability of unusual updates, as well as for detecting novelties. This measure was successfully tested in various scenarios with (i) synthetic data, (ii) real time series datasets, and multiple adaptive filters and learning algorithms. The results of these experiments are presented. |
| topic |
novelty detection learning system learning time series learning entropy extreme seeking entropy |
| url |
https://www.mdpi.com/1099-4300/22/1/93 |
| work_keys_str_mv |
AT janvrba introductiontoextremeseekingentropy AT janmares introductiontoextremeseekingentropy |
| _version_ |
1725176322040266752 |