Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

Adrian Rubio-Solis; Uriel Martinez Hernandez; George Panoutsos

doi:10.1109/FUZZ-IEEE.2018.8491583

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

Adrian Rubio-Solis, Uriel Martinez Hernandez, George Panoutsos

University of Sheffield

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

98 Downloads (Pure)

Abstract

It has been demonstrated that Evolutionary Extreme Learning Machine (E-ELM) is frequently much more efficient than traditional gradient-based algorithms for the
parameter identification of feedforward neural networks. In particular, E-ELM is usually faster and provides a higher tradeoff between accuracy and model simplicity. For that reason, this paper shows that an E-ELM that is based on Particle Swarm Optimisation (PSO) and Extreme Learning machine (ELM) can be extended to the Interval Type-2 Radial Basis Function Neural Network (IT2-RBFNN) with a Karnik-Mendel type-reduction layer. To evaluate the efficiency of E-ELM, the IT2-RBFNN is used as an Interval Type-2 Fuzzy Logic System (IT2 FLS) for the modelling of two popular data sets and for the prediction of chaotic time series. According to our results, E-ELM applied to the IT2-RBFNN not only outperforms adaptive-gradient-based algorithms and provide a better generalisation compared to other existing IT2 fuzzy methodologies, but similarly to pure
fuzzy models, the IT2-RBFNN is also able to preserve some model interpretation and transparency.

Original language	English
Title of host publication	IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)
Publisher	IEEE
Number of pages	8
ISBN (Print)	978-1-5090-6020-7
DOIs	https://doi.org/10.1109/FUZZ-IEEE.2018.8491583
Publication status	Published - 15 Oct 2018

Access to Document

10.1109/FUZZ-IEEE.2018.8491583

hierarchical-cognitive-architecture.PDF
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
Accepted author manuscript, 418 KB

Cite this

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach. / Rubio-Solis, Adrian; Martinez Hernandez, Uriel; Panoutsos, George.
IEEE International Conference on Fuzzy Systems (FUZZ-IEEE). IEEE, 2018. 8491583.

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

@inproceedings{509c65673ce2484a8b0026e6dd461b56,

title = "Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach",

abstract = "It has been demonstrated that Evolutionary Extreme Learning Machine (E-ELM) is frequently much more efficient than traditional gradient-based algorithms for theparameter identification of feedforward neural networks. In particular, E-ELM is usually faster and provides a higher tradeoff between accuracy and model simplicity. For that reason, this paper shows that an E-ELM that is based on Particle Swarm Optimisation (PSO) and Extreme Learning machine (ELM) can be extended to the Interval Type-2 Radial Basis Function Neural Network (IT2-RBFNN) with a Karnik-Mendel type-reduction layer. To evaluate the efficiency of E-ELM, the IT2-RBFNN is used as an Interval Type-2 Fuzzy Logic System (IT2 FLS) for the modelling of two popular data sets and for the prediction of chaotic time series. According to our results, E-ELM applied to the IT2-RBFNN not only outperforms adaptive-gradient-based algorithms and provide a better generalisation compared to other existing IT2 fuzzy methodologies, but similarly to purefuzzy models, the IT2-RBFNN is also able to preserve some model interpretation and transparency.",

author = "Adrian Rubio-Solis and {Martinez Hernandez}, Uriel and George Panoutsos",

year = "2018",

month = oct,

day = "15",

doi = "10.1109/FUZZ-IEEE.2018.8491583",

language = "English",

isbn = "978-1-5090-6020-7",

booktitle = "IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)",

publisher = "IEEE",

address = "USA United States",

}

TY - GEN

T1 - Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

AU - Rubio-Solis, Adrian

AU - Martinez Hernandez, Uriel

AU - Panoutsos, George

PY - 2018/10/15

Y1 - 2018/10/15

N2 - It has been demonstrated that Evolutionary Extreme Learning Machine (E-ELM) is frequently much more efficient than traditional gradient-based algorithms for theparameter identification of feedforward neural networks. In particular, E-ELM is usually faster and provides a higher tradeoff between accuracy and model simplicity. For that reason, this paper shows that an E-ELM that is based on Particle Swarm Optimisation (PSO) and Extreme Learning machine (ELM) can be extended to the Interval Type-2 Radial Basis Function Neural Network (IT2-RBFNN) with a Karnik-Mendel type-reduction layer. To evaluate the efficiency of E-ELM, the IT2-RBFNN is used as an Interval Type-2 Fuzzy Logic System (IT2 FLS) for the modelling of two popular data sets and for the prediction of chaotic time series. According to our results, E-ELM applied to the IT2-RBFNN not only outperforms adaptive-gradient-based algorithms and provide a better generalisation compared to other existing IT2 fuzzy methodologies, but similarly to purefuzzy models, the IT2-RBFNN is also able to preserve some model interpretation and transparency.

AB - It has been demonstrated that Evolutionary Extreme Learning Machine (E-ELM) is frequently much more efficient than traditional gradient-based algorithms for theparameter identification of feedforward neural networks. In particular, E-ELM is usually faster and provides a higher tradeoff between accuracy and model simplicity. For that reason, this paper shows that an E-ELM that is based on Particle Swarm Optimisation (PSO) and Extreme Learning machine (ELM) can be extended to the Interval Type-2 Radial Basis Function Neural Network (IT2-RBFNN) with a Karnik-Mendel type-reduction layer. To evaluate the efficiency of E-ELM, the IT2-RBFNN is used as an Interval Type-2 Fuzzy Logic System (IT2 FLS) for the modelling of two popular data sets and for the prediction of chaotic time series. According to our results, E-ELM applied to the IT2-RBFNN not only outperforms adaptive-gradient-based algorithms and provide a better generalisation compared to other existing IT2 fuzzy methodologies, but similarly to purefuzzy models, the IT2-RBFNN is also able to preserve some model interpretation and transparency.

U2 - 10.1109/FUZZ-IEEE.2018.8491583

DO - 10.1109/FUZZ-IEEE.2018.8491583

M3 - Chapter in a published conference proceeding

SN - 978-1-5090-6020-7

BT - IEEE International Conference on Fuzzy Systems (FUZZ-IEEE)

PB - IEEE

ER -

Evolutionary Extreme Learning Machine for the Interval Type-2 Radial Basis Function Neural Network: A Fuzzy Modelling Approach

Abstract

Access to Document

Fingerprint

Cite this