Conv-Adapter: Exploring Parameter Efficient Transfer Learning for ConvNets
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2022
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911836618620928 |
|---|---|
| author | Chen, Hao Tao, Ran Zhang, Han Wang, Yidong Li, Xiang Ye, Wei Wang, Jindong Hu, Guosheng Savvides, Marios |
| author_facet | Chen, Hao Tao, Ran Zhang, Han Wang, Yidong Li, Xiang Ye, Wei Wang, Jindong Hu, Guosheng Savvides, Marios |
| contents | While parameter efficient tuning (PET) methods have shown great potential with transformer architecture on Natural Language Processing (NLP) tasks, their effectiveness with large-scale ConvNets is still under-studied on Computer Vision (CV) tasks. This paper proposes Conv-Adapter, a PET module designed for ConvNets. Conv-Adapter is light-weight, domain-transferable, and architecture-agnostic with generalized performance on different tasks. When transferring on downstream tasks, Conv-Adapter learns tasks-specific feature modulation to the intermediate representations of backbones while keeping the pre-trained parameters frozen. By introducing only a tiny amount of learnable parameters, e.g., only 3.5% full fine-tuning parameters of ResNet50. It can also be applied for transformer-based backbones. Conv-Adapter outperforms previous PET baseline methods and achieves comparable or surpasses the performance of full fine-tuning on 23 classification tasks of various domains. It also presents superior performance on the few-shot classification with an average margin of 3.39%. Beyond classification, Conv-Adapter can generalize to detection and segmentation tasks with more than 50% reduction of parameters but comparable performance to the traditional full fine-tuning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2208_07463 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Conv-Adapter: Exploring Parameter Efficient Transfer Learning for ConvNets Chen, Hao Tao, Ran Zhang, Han Wang, Yidong Li, Xiang Ye, Wei Wang, Jindong Hu, Guosheng Savvides, Marios Computer Vision and Pattern Recognition Artificial Intelligence While parameter efficient tuning (PET) methods have shown great potential with transformer architecture on Natural Language Processing (NLP) tasks, their effectiveness with large-scale ConvNets is still under-studied on Computer Vision (CV) tasks. This paper proposes Conv-Adapter, a PET module designed for ConvNets. Conv-Adapter is light-weight, domain-transferable, and architecture-agnostic with generalized performance on different tasks. When transferring on downstream tasks, Conv-Adapter learns tasks-specific feature modulation to the intermediate representations of backbones while keeping the pre-trained parameters frozen. By introducing only a tiny amount of learnable parameters, e.g., only 3.5% full fine-tuning parameters of ResNet50. It can also be applied for transformer-based backbones. Conv-Adapter outperforms previous PET baseline methods and achieves comparable or surpasses the performance of full fine-tuning on 23 classification tasks of various domains. It also presents superior performance on the few-shot classification with an average margin of 3.39%. Beyond classification, Conv-Adapter can generalize to detection and segmentation tasks with more than 50% reduction of parameters but comparable performance to the traditional full fine-tuning. |
| title | Conv-Adapter: Exploring Parameter Efficient Transfer Learning for ConvNets |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2208.07463 |