Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection
Image outlier detection has been an important research issue for many computer vision tasks. However, most existing outlier detection methods fail in the high-dimensional image datasets. In order to address this problem, we propose a novel image outlier detection method by combining autoencoder with...
| Main Authors: | , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/104700 http://hdl.handle.net/10220/50296 |
| _version_ | 1811687442353225728 |
|---|---|
| author | Chen, Zhaomin Yeo, Chai Kiat Lee, Bu Sung Lau, Chiew Tong Jin, Yaochu |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Chen, Zhaomin Yeo, Chai Kiat Lee, Bu Sung Lau, Chiew Tong Jin, Yaochu |
| author_sort | Chen, Zhaomin |
| collection | NTU |
| description | Image outlier detection has been an important research issue for many computer vision tasks. However, most existing outlier detection methods fail in the high-dimensional image datasets. In order to address this problem, we propose a novel image outlier detection method by combining autoencoder with Adaboost (ADAE). By ensembling many weak autoencoders, our method can better capture the statistical correlations among the features of normal data than the single autoencoder. Therefore, the proposed ADAE is able to determine the outliers efficiently. In order to reduce the many parameters in ADAE, we introduce the Sparse Group Lasso (SGL) constraint into the learning objective of ADAE. We combine Adagrad with Proximal Gradient Descent to optimize this additional learning objective. We also propose the multi-objective evolutionary algorithm to determine the best penalty factors of SGL. By evaluating on several famous image datasets, the detection results testify to the outstanding outlier detection performance of ADAE. The evaluation results also show SGL can make the detection model more compact while maintaining the similar detection performance. |
| first_indexed | 2024-10-01T05:16:23Z |
| format | Journal Article |
| id | ntu-10356/104700 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T05:16:23Z |
| publishDate | 2019 |
| record_format | dspace |
| spelling | ntu-10356/1047002020-03-07T11:50:39Z Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection Chen, Zhaomin Yeo, Chai Kiat Lee, Bu Sung Lau, Chiew Tong Jin, Yaochu School of Computer Science and Engineering Computer Network and Communication Graduate Lab Engineering::Computer science and engineering Image Outlier Detection Autoencoder Image outlier detection has been an important research issue for many computer vision tasks. However, most existing outlier detection methods fail in the high-dimensional image datasets. In order to address this problem, we propose a novel image outlier detection method by combining autoencoder with Adaboost (ADAE). By ensembling many weak autoencoders, our method can better capture the statistical correlations among the features of normal data than the single autoencoder. Therefore, the proposed ADAE is able to determine the outliers efficiently. In order to reduce the many parameters in ADAE, we introduce the Sparse Group Lasso (SGL) constraint into the learning objective of ADAE. We combine Adagrad with Proximal Gradient Descent to optimize this additional learning objective. We also propose the multi-objective evolutionary algorithm to determine the best penalty factors of SGL. By evaluating on several famous image datasets, the detection results testify to the outstanding outlier detection performance of ADAE. The evaluation results also show SGL can make the detection model more compact while maintaining the similar detection performance. Accepted version 2019-10-31T02:30:43Z 2019-12-06T21:37:50Z 2019-10-31T02:30:43Z 2019-12-06T21:37:50Z 2018 Journal Article Chen, Z., Yeo, C. K., Lee, B. S., Lau, C. T., & Jin, Y. (2018). Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection. Neurocomputing, 309, 192-200. doi:10.1016/j.neucom.2018.05.012 0925-2312 https://hdl.handle.net/10356/104700 http://hdl.handle.net/10220/50296 10.1016/j.neucom.2018.05.012 en Neurocomputing © 2018 Elsevier B.V. All rights reserved. This paper was published in Neurocomputing and is made available with permission of Elsevier B.V. 27 p. application/pdf |
| spellingShingle | Engineering::Computer science and engineering Image Outlier Detection Autoencoder Chen, Zhaomin Yeo, Chai Kiat Lee, Bu Sung Lau, Chiew Tong Jin, Yaochu Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection |
| title | Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection |
| title_full | Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection |
| title_fullStr | Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection |
| title_full_unstemmed | Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection |
| title_short | Evolutionary multi-objective optimization based ensemble autoencoders for image outlier detection |
| title_sort | evolutionary multi objective optimization based ensemble autoencoders for image outlier detection |
| topic | Engineering::Computer science and engineering Image Outlier Detection Autoencoder |
| url | https://hdl.handle.net/10356/104700 http://hdl.handle.net/10220/50296 |
| work_keys_str_mv | AT chenzhaomin evolutionarymultiobjectiveoptimizationbasedensembleautoencodersforimageoutlierdetection AT yeochaikiat evolutionarymultiobjectiveoptimizationbasedensembleautoencodersforimageoutlierdetection AT leebusung evolutionarymultiobjectiveoptimizationbasedensembleautoencodersforimageoutlierdetection AT lauchiewtong evolutionarymultiobjectiveoptimizationbasedensembleautoencodersforimageoutlierdetection AT jinyaochu evolutionarymultiobjectiveoptimizationbasedensembleautoencodersforimageoutlierdetection |