Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification
As a novel neural network learning framework, Twin Extreme Learning Machine (TELM) has received extensive attention and research in the field of machine learning. However, TELM is affected by noise or outliers in practical applications so that its generalization performance is reduced compared to ro...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-08-01
|
| Series: | Axioms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1680/12/8/765 |
| _version_ | 1797585519106850816 |
|---|---|
| author | Yimeng Jiang Guolin Yu Jun Ma |
| author_facet | Yimeng Jiang Guolin Yu Jun Ma |
| author_sort | Yimeng Jiang |
| collection | DOAJ |
| description | As a novel neural network learning framework, Twin Extreme Learning Machine (TELM) has received extensive attention and research in the field of machine learning. However, TELM is affected by noise or outliers in practical applications so that its generalization performance is reduced compared to robust learning algorithms. In this paper, we propose two novel distance metric optimization-driven robust twin extreme learning machine learning frameworks for pattern classification, namely, CWTELM and FCWTELM. By introducing the robust Welsch loss function and capped <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>L</mi><mrow><mn>2</mn><mo>,</mo><mi>p</mi></mrow></msub></semantics></math></inline-formula>-distance metric, our methods reduce the effect of outliers and improve the generalization performance of the model compared to TELM. In addition, two efficient iterative algorithms are designed to solve the challenges brought by the non-convex optimization problems CWTELM and FCWTELM, and we theoretically guarantee their convergence, local optimality, and computational complexity. Then, the proposed algorithms are compared with five other classical algorithms under different noise and different datasets, and the statistical detection analysis is implemented. Finally, we conclude that our algorithm has excellent robustness and classification performance. |
| first_indexed | 2024-03-11T00:07:16Z |
| format | Article |
| id | doaj.art-3a4eadd3838944bc9ce3d2edee314e77 |
| institution | Directory Open Access Journal |
| issn | 2075-1680 |
| language | English |
| last_indexed | 2024-03-11T00:07:16Z |
| publishDate | 2023-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Axioms |
| spelling | doaj.art-3a4eadd3838944bc9ce3d2edee314e772023-11-19T00:14:55ZengMDPI AGAxioms2075-16802023-08-0112876510.3390/axioms12080765Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern ClassificationYimeng Jiang0Guolin Yu1Jun Ma2School of Mathematics and Information Sciences, North Minzu University, Yinchuan 750021, ChinaSchool of Mathematics and Information Sciences, North Minzu University, Yinchuan 750021, ChinaSchool of Mathematics and Information Sciences, North Minzu University, Yinchuan 750021, ChinaAs a novel neural network learning framework, Twin Extreme Learning Machine (TELM) has received extensive attention and research in the field of machine learning. However, TELM is affected by noise or outliers in practical applications so that its generalization performance is reduced compared to robust learning algorithms. In this paper, we propose two novel distance metric optimization-driven robust twin extreme learning machine learning frameworks for pattern classification, namely, CWTELM and FCWTELM. By introducing the robust Welsch loss function and capped <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msub><mi>L</mi><mrow><mn>2</mn><mo>,</mo><mi>p</mi></mrow></msub></semantics></math></inline-formula>-distance metric, our methods reduce the effect of outliers and improve the generalization performance of the model compared to TELM. In addition, two efficient iterative algorithms are designed to solve the challenges brought by the non-convex optimization problems CWTELM and FCWTELM, and we theoretically guarantee their convergence, local optimality, and computational complexity. Then, the proposed algorithms are compared with five other classical algorithms under different noise and different datasets, and the statistical detection analysis is implemented. Finally, we conclude that our algorithm has excellent robustness and classification performance.https://www.mdpi.com/2075-1680/12/8/765neural networktwin extreme learning machinedistance metricrobustnesspattern classification |
| spellingShingle | Yimeng Jiang Guolin Yu Jun Ma Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification Axioms neural network twin extreme learning machine distance metric robustness pattern classification |
| title | Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification |
| title_full | Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification |
| title_fullStr | Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification |
| title_full_unstemmed | Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification |
| title_short | Distance Metric Optimization-Driven Neural Network Learning Framework for Pattern Classification |
| title_sort | distance metric optimization driven neural network learning framework for pattern classification |
| topic | neural network twin extreme learning machine distance metric robustness pattern classification |
| url | https://www.mdpi.com/2075-1680/12/8/765 |
| work_keys_str_mv | AT yimengjiang distancemetricoptimizationdrivenneuralnetworklearningframeworkforpatternclassification AT guolinyu distancemetricoptimizationdrivenneuralnetworklearningframeworkforpatternclassification AT junma distancemetricoptimizationdrivenneuralnetworklearningframeworkforpatternclassification |