FSL-Rectifier: Rectify Outliers in Few-Shot Learning via Test-Time Augmentation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913617393221632 |
|---|---|
| author | Bai, Yunwei Tan, Ying Kiat Chen, Shiming Shu, Yao Chen, Tsuhan |
| author_facet | Bai, Yunwei Tan, Ying Kiat Chen, Shiming Shu, Yao Chen, Tsuhan |
| contents | Few-shot learning (FSL) commonly requires a model to identify images (queries) that belong to classes unseen during training, based on a few labelled samples of the new classes (support set) as reference. So far, plenty of algorithms involve training data augmentation to improve the generalization capability of FSL models, but outlier queries or support images during inference can still pose great generalization challenges. In this work, to reduce the bias caused by the outlier samples, we generate additional test-class samples by combining original samples with suitable train-class samples via a generative image combiner. Then, we obtain averaged features via an augmentor, which leads to more typical representations through the averaging. We experimentally and theoretically demonstrate the effectiveness of our method, obtaining a test accuracy improvement proportion of around 10\% (e.g., from 46.86\% to 53.28\%) for trained FSL models. Importantly, given a pretrained image combiner, our method is training-free for off-the-shelf FSL models, whose performance can be improved without extra datasets nor further training of the models themselves. Codes are available at https://github.com/WendyBaiYunwei/FSL-Rectifier-Pub. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_18292 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | FSL-Rectifier: Rectify Outliers in Few-Shot Learning via Test-Time Augmentation Bai, Yunwei Tan, Ying Kiat Chen, Shiming Shu, Yao Chen, Tsuhan Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Few-shot learning (FSL) commonly requires a model to identify images (queries) that belong to classes unseen during training, based on a few labelled samples of the new classes (support set) as reference. So far, plenty of algorithms involve training data augmentation to improve the generalization capability of FSL models, but outlier queries or support images during inference can still pose great generalization challenges. In this work, to reduce the bias caused by the outlier samples, we generate additional test-class samples by combining original samples with suitable train-class samples via a generative image combiner. Then, we obtain averaged features via an augmentor, which leads to more typical representations through the averaging. We experimentally and theoretically demonstrate the effectiveness of our method, obtaining a test accuracy improvement proportion of around 10\% (e.g., from 46.86\% to 53.28\%) for trained FSL models. Importantly, given a pretrained image combiner, our method is training-free for off-the-shelf FSL models, whose performance can be improved without extra datasets nor further training of the models themselves. Codes are available at https://github.com/WendyBaiYunwei/FSL-Rectifier-Pub. |
| title | FSL-Rectifier: Rectify Outliers in Few-Shot Learning via Test-Time Augmentation |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2402.18292 |