On the Effectiveness of Supervision in Asymmetric Non-Contrastive Learning
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_ | 1866917695265439744 |
|---|---|
| author | Oh, Jeongheon Lee, Kibok |
| author_facet | Oh, Jeongheon Lee, Kibok |
| contents | Supervised contrastive representation learning has been shown to be effective in various transfer learning scenarios. However, while asymmetric non-contrastive learning (ANCL) often outperforms its contrastive learning counterpart in self-supervised representation learning, the extension of ANCL to supervised scenarios is less explored. To bridge the gap, we study ANCL for supervised representation learning, coined SupSiam and SupBYOL, leveraging labels in ANCL to achieve better representations. The proposed supervised ANCL framework improves representation learning while avoiding collapse. Our analysis reveals that providing supervision to ANCL reduces intra-class variance, and the contribution of supervision should be adjusted to achieve the best performance. Experiments demonstrate the superiority of supervised ANCL across various datasets and tasks. The code is available at: https://github.com/JH-Oh-23/Sup-ANCL. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_10815 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | On the Effectiveness of Supervision in Asymmetric Non-Contrastive Learning Oh, Jeongheon Lee, Kibok Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition Supervised contrastive representation learning has been shown to be effective in various transfer learning scenarios. However, while asymmetric non-contrastive learning (ANCL) often outperforms its contrastive learning counterpart in self-supervised representation learning, the extension of ANCL to supervised scenarios is less explored. To bridge the gap, we study ANCL for supervised representation learning, coined SupSiam and SupBYOL, leveraging labels in ANCL to achieve better representations. The proposed supervised ANCL framework improves representation learning while avoiding collapse. Our analysis reveals that providing supervision to ANCL reduces intra-class variance, and the contribution of supervision should be adjusted to achieve the best performance. Experiments demonstrate the superiority of supervised ANCL across various datasets and tasks. The code is available at: https://github.com/JH-Oh-23/Sup-ANCL. |
| title | On the Effectiveness of Supervision in Asymmetric Non-Contrastive Learning |
| topic | Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2406.10815 |