An Interpretable Compression and Classification System: Theory and Applications
This study proposes a low-complexity interpretable classification system. The proposed system contains main modules including feature extraction, feature reduction, and classification. All of them are linear. Thanks to the linear property, the extracted and reduced features can be inversed to origin...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9159554/ |
| _version_ | 1818936298677731328 |
|---|---|
| author | Tzu-Wei Tseng Kai-Jiun Yang C.-C. Jay Kuo Shang-Ho Tsai |
| author_facet | Tzu-Wei Tseng Kai-Jiun Yang C.-C. Jay Kuo Shang-Ho Tsai |
| author_sort | Tzu-Wei Tseng |
| collection | DOAJ |
| description | This study proposes a low-complexity interpretable classification system. The proposed system contains main modules including feature extraction, feature reduction, and classification. All of them are linear. Thanks to the linear property, the extracted and reduced features can be inversed to original data, like a linear transform such as Fourier transform, so that one can quantify and visualize the contribution of individual features towards the original data. Also, the reduced features and reversibility naturally endure the proposed system ability of data compression. This system can significantly compress data with a small percent deviation between the compressed and the original data. At the same time, when the compressed data is used for classification, it still achieves high testing accuracy. Furthermore, we observe that the extracted features of the proposed system can be approximated to uncorrelated Gaussian random variables. Hence, classical theory in estimation and detection can be applied for classification. This motivates us to propose using a MAP (maximum a posteriori) based classification method. As a result, the extracted features and the corresponding performance have statistical meaning and mathematically interpretable. Simulation results show that the proposed classification system not only enjoys significant reduced training and testing time but also high testing accuracy compared to the conventional schemes. |
| first_indexed | 2024-12-20T05:33:51Z |
| format | Article |
| id | doaj.art-a2c5c19bb3bf48dbad3fae9c6f059f4f |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-20T05:33:51Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-a2c5c19bb3bf48dbad3fae9c6f059f4f2022-12-21T19:51:41ZengIEEEIEEE Access2169-35362020-01-01814396214397410.1109/ACCESS.2020.30143079159554An Interpretable Compression and Classification System: Theory and ApplicationsTzu-Wei Tseng0https://orcid.org/0000-0002-3229-0212Kai-Jiun Yang1https://orcid.org/0000-0002-6562-5058C.-C. Jay Kuo2https://orcid.org/0000-0001-9474-5035Shang-Ho Tsai3https://orcid.org/0000-0001-9055-6333Department of Electrical Engineering, National Chiao Tung University, Hsinchu, TaiwanSystem Integration & Applications Department, Division for Embedded System & SoC Technology Information, Communications Research Laboratories, Industrial Technology Research Institute, Zhudong, TaiwanDepartment of Electrical Engineering, Signal and Image Processing Institute, University of Southern California, Los Angeles, CA, USADepartment of Electrical Engineering, National Chiao Tung University, Hsinchu, TaiwanThis study proposes a low-complexity interpretable classification system. The proposed system contains main modules including feature extraction, feature reduction, and classification. All of them are linear. Thanks to the linear property, the extracted and reduced features can be inversed to original data, like a linear transform such as Fourier transform, so that one can quantify and visualize the contribution of individual features towards the original data. Also, the reduced features and reversibility naturally endure the proposed system ability of data compression. This system can significantly compress data with a small percent deviation between the compressed and the original data. At the same time, when the compressed data is used for classification, it still achieves high testing accuracy. Furthermore, we observe that the extracted features of the proposed system can be approximated to uncorrelated Gaussian random variables. Hence, classical theory in estimation and detection can be applied for classification. This motivates us to propose using a MAP (maximum a posteriori) based classification method. As a result, the extracted features and the corresponding performance have statistical meaning and mathematically interpretable. Simulation results show that the proposed classification system not only enjoys significant reduced training and testing time but also high testing accuracy compared to the conventional schemes.https://ieeexplore.ieee.org/document/9159554/Classificationconvolution neural networkdata compressionfeature extractionfeature reductionimage recognition |
| spellingShingle | Tzu-Wei Tseng Kai-Jiun Yang C.-C. Jay Kuo Shang-Ho Tsai An Interpretable Compression and Classification System: Theory and Applications IEEE Access Classification convolution neural network data compression feature extraction feature reduction image recognition |
| title | An Interpretable Compression and Classification System: Theory and Applications |
| title_full | An Interpretable Compression and Classification System: Theory and Applications |
| title_fullStr | An Interpretable Compression and Classification System: Theory and Applications |
| title_full_unstemmed | An Interpretable Compression and Classification System: Theory and Applications |
| title_short | An Interpretable Compression and Classification System: Theory and Applications |
| title_sort | interpretable compression and classification system theory and applications |
| topic | Classification convolution neural network data compression feature extraction feature reduction image recognition |
| url | https://ieeexplore.ieee.org/document/9159554/ |
| work_keys_str_mv | AT tzuweitseng aninterpretablecompressionandclassificationsystemtheoryandapplications AT kaijiunyang aninterpretablecompressionandclassificationsystemtheoryandapplications AT ccjaykuo aninterpretablecompressionandclassificationsystemtheoryandapplications AT shanghotsai aninterpretablecompressionandclassificationsystemtheoryandapplications AT tzuweitseng interpretablecompressionandclassificationsystemtheoryandapplications AT kaijiunyang interpretablecompressionandclassificationsystemtheoryandapplications AT ccjaykuo interpretablecompressionandclassificationsystemtheoryandapplications AT shanghotsai interpretablecompressionandclassificationsystemtheoryandapplications |