Retro: Reusing teacher projection head for efficient embedding distillation on Lightweight Models via Self-supervised 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_ | 1866929472215711744 |
|---|---|
| author | Nguyen, Khanh-Binh Park, Chae Jung |
| author_facet | Nguyen, Khanh-Binh Park, Chae Jung |
| contents | Self-supervised learning (SSL) is gaining attention for its ability to learn effective representations with large amounts of unlabeled data. Lightweight models can be distilled from larger self-supervised pre-trained models using contrastive and consistency constraints. Still, the different sizes of the projection heads make it challenging for students to mimic the teacher's embedding accurately. We propose \textsc{Retro}, which reuses the teacher's projection head for students, and our experimental results demonstrate significant improvements over the state-of-the-art on all lightweight models. For instance, when training EfficientNet-B0 using ResNet-50/101/152 as teachers, our approach improves the linear result on ImageNet to $66.9\%$, $69.3\%$, and $69.8\%$, respectively, with significantly fewer parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_15311 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Retro: Reusing teacher projection head for efficient embedding distillation on Lightweight Models via Self-supervised Learning Nguyen, Khanh-Binh Park, Chae Jung Computer Vision and Pattern Recognition Artificial Intelligence Self-supervised learning (SSL) is gaining attention for its ability to learn effective representations with large amounts of unlabeled data. Lightweight models can be distilled from larger self-supervised pre-trained models using contrastive and consistency constraints. Still, the different sizes of the projection heads make it challenging for students to mimic the teacher's embedding accurately. We propose \textsc{Retro}, which reuses the teacher's projection head for students, and our experimental results demonstrate significant improvements over the state-of-the-art on all lightweight models. For instance, when training EfficientNet-B0 using ResNet-50/101/152 as teachers, our approach improves the linear result on ImageNet to $66.9\%$, $69.3\%$, and $69.8\%$, respectively, with significantly fewer parameters. |
| title | Retro: Reusing teacher projection head for efficient embedding distillation on Lightweight Models via Self-supervised Learning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2405.15311 |