LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things
Edge-based intelligent data analytics supports the Industrial Internet of Things (IIoT) to enable efficient manufacturing. Incremental learning in the edge-based data analytics has the potential to analyze continuously collected real-time data. However, additional efforts are needed to address perfo...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10343149/ |
| _version_ | 1797376260609933312 |
|---|---|
| author | Atallo Kassaw Takele Balazs Villanyi |
| author_facet | Atallo Kassaw Takele Balazs Villanyi |
| author_sort | Atallo Kassaw Takele |
| collection | DOAJ |
| description | Edge-based intelligent data analytics supports the Industrial Internet of Things (IIoT) to enable efficient manufacturing. Incremental learning in the edge-based data analytics has the potential to analyze continuously collected real-time data. However, additional efforts are needed to address performance, latency, resource utilization and storage of historical data challenges. This paper introduces an incremental learning approach based on Long-Short Term Memory (LSTM) autoencoders, by sparsening the weight matrix and taking samples from previously trained sub-datasets. The aim is to minimize the resources utilized while training redundant knowledge for edge devices of IIoT. The degree of sparsity can be determined by the redundancy of patterns, and the inverse of the coefficient of variation has been utilized to recognize it. A higher value of the inverse of the coefficient of variation shows that the values of the weight matrix are close to each other, which indicates the redundancy of knowledge, and vice versa. In addition, the coefficient of variation has been applied for limiting the size of samples from the previously trained sub-datasets. The experiment conducted using the IIoT testbed dataset demonstrates substantial enhancements in resource optimization without compromising performance. |
| first_indexed | 2024-03-08T19:36:56Z |
| format | Article |
| id | doaj.art-04ced91100a6435c88ea62e12bed561e |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-08T19:36:56Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-04ced91100a6435c88ea62e12bed561e2023-12-26T00:08:37ZengIEEEIEEE Access2169-35362023-01-011113792913793610.1109/ACCESS.2023.333955610343149LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of ThingsAtallo Kassaw Takele0https://orcid.org/0000-0002-6679-7679Balazs Villanyi1https://orcid.org/0000-0003-2873-9934Department of Electronics Technology, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Budapest, HungaryDepartment of Electronics Technology, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Budapest, HungaryEdge-based intelligent data analytics supports the Industrial Internet of Things (IIoT) to enable efficient manufacturing. Incremental learning in the edge-based data analytics has the potential to analyze continuously collected real-time data. However, additional efforts are needed to address performance, latency, resource utilization and storage of historical data challenges. This paper introduces an incremental learning approach based on Long-Short Term Memory (LSTM) autoencoders, by sparsening the weight matrix and taking samples from previously trained sub-datasets. The aim is to minimize the resources utilized while training redundant knowledge for edge devices of IIoT. The degree of sparsity can be determined by the redundancy of patterns, and the inverse of the coefficient of variation has been utilized to recognize it. A higher value of the inverse of the coefficient of variation shows that the values of the weight matrix are close to each other, which indicates the redundancy of knowledge, and vice versa. In addition, the coefficient of variation has been applied for limiting the size of samples from the previously trained sub-datasets. The experiment conducted using the IIoT testbed dataset demonstrates substantial enhancements in resource optimization without compromising performance.https://ieeexplore.ieee.org/document/10343149/Industrial internet of things (IIoT)incremental learningLSTM-autoencoderweight sparsification |
| spellingShingle | Atallo Kassaw Takele Balazs Villanyi LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things IEEE Access Industrial internet of things (IIoT) incremental learning LSTM-autoencoder weight sparsification |
| title | LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things |
| title_full | LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things |
| title_fullStr | LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things |
| title_full_unstemmed | LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things |
| title_short | LSTM-Autoencoder-Based Incremental Learning for Industrial Internet of Things |
| title_sort | lstm autoencoder based incremental learning for industrial internet of things |
| topic | Industrial internet of things (IIoT) incremental learning LSTM-autoencoder weight sparsification |
| url | https://ieeexplore.ieee.org/document/10343149/ |
| work_keys_str_mv | AT atallokassawtakele lstmautoencoderbasedincrementallearningforindustrialinternetofthings AT balazsvillanyi lstmautoencoderbasedincrementallearningforindustrialinternetofthings |