Home > Research > Publications & Outputs > Multilayer Evolving Fuzzy Neural Networks with ...

Electronic data

  • Author accepted final version

    Accepted author manuscript, 2.11 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Multilayer Evolving Fuzzy Neural Networks with Self-Adaptive Dimensionality Compression for High-Dimensional Data Classification

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Multilayer Evolving Fuzzy Neural Networks with Self-Adaptive Dimensionality Compression for High-Dimensional Data Classification. / Gu, Xiaowei; Ni, Qiang; Shen, Qiang.
In: IEEE Transactions on Fuzzy Systems, 21.08.2024.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Gu X, Ni Q, Shen Q. Multilayer Evolving Fuzzy Neural Networks with Self-Adaptive Dimensionality Compression for High-Dimensional Data Classification. IEEE Transactions on Fuzzy Systems. 2024 Aug 21. Epub 2024 Aug 21. doi: 10.1109/tfuzz.2024.3446959

Author

Bibtex

@article{39706fffe68e4e1f9a5c58e13157404f,
title = "Multilayer Evolving Fuzzy Neural Networks with Self-Adaptive Dimensionality Compression for High-Dimensional Data Classification",
abstract = "High-dimensional data classification is widely considered as a challenging task in machine learning due to the so-called “curse of dimensionality”. In this paper, a novel multilayer jointly evolving and compressing fuzzy neural network (MECFNN) is proposed to learn highly compact multi-level latent representations from high-dimensional data. As a meta-level stacking ensemble system, each layer of MECFNN is based on a single jointly evolving and compressing neural fuzzy inference system (ECNFIS) that self-organises a set of human-interpretable fuzzy rules from input data in a sample-wise manner to perform approximate reasoning. ECNFISs associate a unique compressive projection matrix to each individual fuzzy rule to compress the consequent part into a tighter form, removing redundant information whilst boosting the diversity within the stacking ensemble. The compressive projection matrices of the cascading ECNFISs are self-updating to minimise the prediction errors via error backpropagation together with the consequent parameters, empowering MECFNN to learn more meaningful, discriminative representations from data at multiple levels of abstraction. An adaptive activation control scheme is further introduced in MECFNN to dynamically exclude less activated fuzzy rules, effectively reducing the computational complexity and fostering generalisation. Numerical examples on popular high-dimensional classification problems demonstrate the efficacy of MECFNN.",
author = "Xiaowei Gu and Qiang Ni and Qiang Shen",
year = "2024",
month = aug,
day = "21",
doi = "10.1109/tfuzz.2024.3446959",
language = "English",
journal = "IEEE Transactions on Fuzzy Systems",
issn = "1063-6706",
publisher = "IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC",

}

RIS

TY - JOUR

T1 - Multilayer Evolving Fuzzy Neural Networks with Self-Adaptive Dimensionality Compression for High-Dimensional Data Classification

AU - Gu, Xiaowei

AU - Ni, Qiang

AU - Shen, Qiang

PY - 2024/8/21

Y1 - 2024/8/21

N2 - High-dimensional data classification is widely considered as a challenging task in machine learning due to the so-called “curse of dimensionality”. In this paper, a novel multilayer jointly evolving and compressing fuzzy neural network (MECFNN) is proposed to learn highly compact multi-level latent representations from high-dimensional data. As a meta-level stacking ensemble system, each layer of MECFNN is based on a single jointly evolving and compressing neural fuzzy inference system (ECNFIS) that self-organises a set of human-interpretable fuzzy rules from input data in a sample-wise manner to perform approximate reasoning. ECNFISs associate a unique compressive projection matrix to each individual fuzzy rule to compress the consequent part into a tighter form, removing redundant information whilst boosting the diversity within the stacking ensemble. The compressive projection matrices of the cascading ECNFISs are self-updating to minimise the prediction errors via error backpropagation together with the consequent parameters, empowering MECFNN to learn more meaningful, discriminative representations from data at multiple levels of abstraction. An adaptive activation control scheme is further introduced in MECFNN to dynamically exclude less activated fuzzy rules, effectively reducing the computational complexity and fostering generalisation. Numerical examples on popular high-dimensional classification problems demonstrate the efficacy of MECFNN.

AB - High-dimensional data classification is widely considered as a challenging task in machine learning due to the so-called “curse of dimensionality”. In this paper, a novel multilayer jointly evolving and compressing fuzzy neural network (MECFNN) is proposed to learn highly compact multi-level latent representations from high-dimensional data. As a meta-level stacking ensemble system, each layer of MECFNN is based on a single jointly evolving and compressing neural fuzzy inference system (ECNFIS) that self-organises a set of human-interpretable fuzzy rules from input data in a sample-wise manner to perform approximate reasoning. ECNFISs associate a unique compressive projection matrix to each individual fuzzy rule to compress the consequent part into a tighter form, removing redundant information whilst boosting the diversity within the stacking ensemble. The compressive projection matrices of the cascading ECNFISs are self-updating to minimise the prediction errors via error backpropagation together with the consequent parameters, empowering MECFNN to learn more meaningful, discriminative representations from data at multiple levels of abstraction. An adaptive activation control scheme is further introduced in MECFNN to dynamically exclude less activated fuzzy rules, effectively reducing the computational complexity and fostering generalisation. Numerical examples on popular high-dimensional classification problems demonstrate the efficacy of MECFNN.

U2 - 10.1109/tfuzz.2024.3446959

DO - 10.1109/tfuzz.2024.3446959

M3 - Journal article

JO - IEEE Transactions on Fuzzy Systems

JF - IEEE Transactions on Fuzzy Systems

SN - 1063-6706

ER -