Correntropy-Based Evolving Fuzzy Neural System
- Submitting institution
-
The University of Lancaster
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 229359498
- Type
- D - Journal article
- DOI
-
10.1109/TFUZZ.2017.2719619
- Title of journal
- IEEE Transactions on Fuzzy Systems
- Article number
- -
- First page
- 1324
- Volume
- 26
- Issue
- 3
- ISSN
- 1063-6706
- Open access status
- Compliant
- Month of publication
- June
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
B - Data Science
- Citation count
- 18
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Self-evolving models are very important for complex dynamic and non-stationary systems. Their design is routinely based on minimisation of root mean square error. However, specifically for such complex and dynamic systems with structural evolution it is of great importance to have a more robust error minimisation criteria. In this paper correntropy is pioneered for this type of systems and proves to bring significantly better results in handling non-Gaussian noise. This paper is a result of a collaboration with a visiting researcher and was instrumental in getting a 1.3M InnovateUK grant CTHULHE and Royal Society exchange grant with University of Florida.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
