LTRL: Boosting Long-tail Recognition via Reflective Learning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914948558356480 |
|---|---|
| author | Zhao, Qihao Dai, Yalun Lin, Shen Hu, Wei Zhang, Fan Liu, Jun |
| author_facet | Zhao, Qihao Dai, Yalun Lin, Shen Hu, Wei Zhang, Fan Liu, Jun |
| contents | In real-world scenarios, where knowledge distributions exhibit long-tail. Humans manage to master knowledge uniformly across imbalanced distributions, a feat attributed to their diligent practices of reviewing, summarizing, and correcting errors. Motivated by this learning process, we propose a novel learning paradigm, called reflecting learning, in handling long-tail recognition. Our method integrates three processes for reviewing past predictions during training, summarizing and leveraging the feature relation across classes, and correcting gradient conflict for loss functions. These designs are lightweight enough to plug and play with existing long-tail learning methods, achieving state-of-the-art performance in popular long-tail visual benchmarks. The experimental results highlight the great potential of reflecting learning in dealing with long-tail recognition. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_12568 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | LTRL: Boosting Long-tail Recognition via Reflective Learning Zhao, Qihao Dai, Yalun Lin, Shen Hu, Wei Zhang, Fan Liu, Jun Computer Vision and Pattern Recognition In real-world scenarios, where knowledge distributions exhibit long-tail. Humans manage to master knowledge uniformly across imbalanced distributions, a feat attributed to their diligent practices of reviewing, summarizing, and correcting errors. Motivated by this learning process, we propose a novel learning paradigm, called reflecting learning, in handling long-tail recognition. Our method integrates three processes for reviewing past predictions during training, summarizing and leveraging the feature relation across classes, and correcting gradient conflict for loss functions. These designs are lightweight enough to plug and play with existing long-tail learning methods, achieving state-of-the-art performance in popular long-tail visual benchmarks. The experimental results highlight the great potential of reflecting learning in dealing with long-tail recognition. |
| title | LTRL: Boosting Long-tail Recognition via Reflective Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.12568 |