Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features
Random Vector Functional Link (RVFL) is popular among researchers in many areas of machine learning. RVFL is preferred by many researchers as RVFL can produce good performance with relatively little training time. Recent works extend RVFL into deep and ensemble versions. However, RVFL does not have...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/160251 |
| _version_ | 1811695829086371840 |
|---|---|
| author | Cheng, Wenxin Suganthan, Ponnuthurai Nagaratnam Katuwal, Rakesh |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Cheng, Wenxin Suganthan, Ponnuthurai Nagaratnam Katuwal, Rakesh |
| author_sort | Cheng, Wenxin |
| collection | NTU |
| description | Random Vector Functional Link (RVFL) is popular among researchers in many areas of machine learning. RVFL is preferred by many researchers as RVFL can produce good performance with relatively little training time. Recent works extend RVFL into deep and ensemble versions. However, RVFL does not have effective feature extraction methods commonly used in time series classification. This results in poor performance of RVFL in time series classification tasks. Also, deep RVFL is a relatively new and evolving area of research. In this paper, we present a framework that extracts features from Residual Networks (Resnet) and trains Ensemble Deep Random Vector Functional Link (edRVFL). We use features extracted from every residual block to train an ensemble of edRVFLs. We propose the following enhancements to edRVFL. Firstly, we diversity the structure of edRVFL and the direct link features to encourage diversity. Secondly, we built an ensemble of edRVFLs with the top two activation functions. Thirdly, we use two-stage tuning to save computational costs. Lastly, we perform a weighted average of all decisions made by every edRVFL. Experiments on the 55 largest UCR datasets show that using features extracted from all Residual blocks improves performance. All our proposed enhancements help improve classification accuracy or computational effort. Consequently, our proposed framework outperforms all traditional and deep learning-based time series classification methods. |
| first_indexed | 2024-10-01T07:29:41Z |
| format | Journal Article |
| id | ntu-10356/160251 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T07:29:41Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | ntu-10356/1602512022-07-18T06:19:10Z Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features Cheng, Wenxin Suganthan, Ponnuthurai Nagaratnam Katuwal, Rakesh School of Electrical and Electronic Engineering Engineering::Electrical and electronic engineering Time Series Classification Ensemble Deep Learning Random Vector Functional Link (RVFL) is popular among researchers in many areas of machine learning. RVFL is preferred by many researchers as RVFL can produce good performance with relatively little training time. Recent works extend RVFL into deep and ensemble versions. However, RVFL does not have effective feature extraction methods commonly used in time series classification. This results in poor performance of RVFL in time series classification tasks. Also, deep RVFL is a relatively new and evolving area of research. In this paper, we present a framework that extracts features from Residual Networks (Resnet) and trains Ensemble Deep Random Vector Functional Link (edRVFL). We use features extracted from every residual block to train an ensemble of edRVFLs. We propose the following enhancements to edRVFL. Firstly, we diversity the structure of edRVFL and the direct link features to encourage diversity. Secondly, we built an ensemble of edRVFLs with the top two activation functions. Thirdly, we use two-stage tuning to save computational costs. Lastly, we perform a weighted average of all decisions made by every edRVFL. Experiments on the 55 largest UCR datasets show that using features extracted from all Residual blocks improves performance. All our proposed enhancements help improve classification accuracy or computational effort. Consequently, our proposed framework outperforms all traditional and deep learning-based time series classification methods. 2022-07-18T06:19:10Z 2022-07-18T06:19:10Z 2021 Journal Article Cheng, W., Suganthan, P. N. & Katuwal, R. (2021). Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features. Applied Soft Computing, 112, 107826-. https://dx.doi.org/10.1016/j.asoc.2021.107826 1568-4946 https://hdl.handle.net/10356/160251 10.1016/j.asoc.2021.107826 2-s2.0-85114239118 112 107826 en Applied Soft Computing © 2021 Elsevier B.V. All rights reserved. |
| spellingShingle | Engineering::Electrical and electronic engineering Time Series Classification Ensemble Deep Learning Cheng, Wenxin Suganthan, Ponnuthurai Nagaratnam Katuwal, Rakesh Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features |
| title | Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features |
| title_full | Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features |
| title_fullStr | Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features |
| title_full_unstemmed | Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features |
| title_short | Time series classification using diversified Ensemble Deep Random Vector Functional Link and Resnet features |
| title_sort | time series classification using diversified ensemble deep random vector functional link and resnet features |
| topic | Engineering::Electrical and electronic engineering Time Series Classification Ensemble Deep Learning |
| url | https://hdl.handle.net/10356/160251 |
| work_keys_str_mv | AT chengwenxin timeseriesclassificationusingdiversifiedensembledeeprandomvectorfunctionallinkandresnetfeatures AT suganthanponnuthurainagaratnam timeseriesclassificationusingdiversifiedensembledeeprandomvectorfunctionallinkandresnetfeatures AT katuwalrakesh timeseriesclassificationusingdiversifiedensembledeeprandomvectorfunctionallinkandresnetfeatures |