A Large-Class Few-Shot Learning Method Based on High-Dimensional Features
Large-class few-shot learning has a wide range of applications in many fields, such as the medical, power, security, and remote sensing fields. At present, many few-shot learning methods for fewer-class scenarios have been proposed, but little research has been performed for large-class scenarios. I...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-11-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/23/12843 |
| _version_ | 1797400418299412480 |
|---|---|
| author | Jiawei Dang Yu Zhou Ruirui Zheng Jianjun He |
| author_facet | Jiawei Dang Yu Zhou Ruirui Zheng Jianjun He |
| author_sort | Jiawei Dang |
| collection | DOAJ |
| description | Large-class few-shot learning has a wide range of applications in many fields, such as the medical, power, security, and remote sensing fields. At present, many few-shot learning methods for fewer-class scenarios have been proposed, but little research has been performed for large-class scenarios. In this paper, we propose a large-class few-shot learning method called HF-FSL, which is based on high-dimensional features. Recent theoretical research shows that if the distribution of samples in a high-dimensional feature space meets the conditions of compactness within the class and the dispersion between classes, the large-class few-shot learning method has a better generalization ability. Inspired by this theory, the basic idea is use a deep neural network to extract high-dimensional features and unitize them to project the samples onto a hypersphere. The global orthogonal regularization strategy can then be used to make samples of different classes on the hypersphere that are as orthogonal as possible, so as to achieve the goal of sample compactness within the class and the dispersion between classes in high-dimensional feature space. Experiments on Omniglot, Fungi, and ImageNet demonstrate that the proposed method can effectively improve the recognition accuracy in a large-class FSL problem. |
| first_indexed | 2024-03-09T01:55:14Z |
| format | Article |
| id | doaj.art-d9bf05d2d39448e7b3693573f97605d6 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T01:55:14Z |
| publishDate | 2023-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-d9bf05d2d39448e7b3693573f97605d62023-12-08T15:11:54ZengMDPI AGApplied Sciences2076-34172023-11-0113231284310.3390/app132312843A Large-Class Few-Shot Learning Method Based on High-Dimensional FeaturesJiawei Dang0Yu Zhou1Ruirui Zheng2Jianjun He3College of Information and Communication Engineering, Dalian Minzu University, Dalian 116620, ChinaCollege of Information and Communication Engineering, Dalian Minzu University, Dalian 116620, ChinaCollege of Information and Communication Engineering, Dalian Minzu University, Dalian 116620, ChinaCollege of Information and Communication Engineering, Dalian Minzu University, Dalian 116620, ChinaLarge-class few-shot learning has a wide range of applications in many fields, such as the medical, power, security, and remote sensing fields. At present, many few-shot learning methods for fewer-class scenarios have been proposed, but little research has been performed for large-class scenarios. In this paper, we propose a large-class few-shot learning method called HF-FSL, which is based on high-dimensional features. Recent theoretical research shows that if the distribution of samples in a high-dimensional feature space meets the conditions of compactness within the class and the dispersion between classes, the large-class few-shot learning method has a better generalization ability. Inspired by this theory, the basic idea is use a deep neural network to extract high-dimensional features and unitize them to project the samples onto a hypersphere. The global orthogonal regularization strategy can then be used to make samples of different classes on the hypersphere that are as orthogonal as possible, so as to achieve the goal of sample compactness within the class and the dispersion between classes in high-dimensional feature space. Experiments on Omniglot, Fungi, and ImageNet demonstrate that the proposed method can effectively improve the recognition accuracy in a large-class FSL problem.https://www.mdpi.com/2076-3417/13/23/12843large-class few-shot learninghigh-dimensional feature spaceglobal orthogonal regularization |
| spellingShingle | Jiawei Dang Yu Zhou Ruirui Zheng Jianjun He A Large-Class Few-Shot Learning Method Based on High-Dimensional Features Applied Sciences large-class few-shot learning high-dimensional feature space global orthogonal regularization |
| title | A Large-Class Few-Shot Learning Method Based on High-Dimensional Features |
| title_full | A Large-Class Few-Shot Learning Method Based on High-Dimensional Features |
| title_fullStr | A Large-Class Few-Shot Learning Method Based on High-Dimensional Features |
| title_full_unstemmed | A Large-Class Few-Shot Learning Method Based on High-Dimensional Features |
| title_short | A Large-Class Few-Shot Learning Method Based on High-Dimensional Features |
| title_sort | large class few shot learning method based on high dimensional features |
| topic | large-class few-shot learning high-dimensional feature space global orthogonal regularization |
| url | https://www.mdpi.com/2076-3417/13/23/12843 |
| work_keys_str_mv | AT jiaweidang alargeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT yuzhou alargeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT ruiruizheng alargeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT jianjunhe alargeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT jiaweidang largeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT yuzhou largeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT ruiruizheng largeclassfewshotlearningmethodbasedonhighdimensionalfeatures AT jianjunhe largeclassfewshotlearningmethodbasedonhighdimensionalfeatures |