2000 character limit reached
Unsupervised Domain Adaptation within Deep Foundation Latent Spaces (2402.14976v1)
Published 22 Feb 2024 in cs.CV, cs.AI, and cs.LG
Abstract: The vision transformer-based foundation models, such as ViT or Dino-V2, are aimed at solving problems with little or no finetuning of features. Using a setting of prototypical networks, we analyse to what extent such foundation models can solve unsupervised domain adaptation without finetuning over the source or target domain. Through quantitative analysis, as well as qualitative interpretations of decision making, we demonstrate that the suggested method can improve upon existing baselines, as well as showcase the limitations of such approach yet to be solved.
- Towards explainable deep neural networks (xdnn). Neural Networks, 130:185–194, 2020.
- Towards interpretable-by-design deep learning algorithms. arXiv preprint arXiv:2311.11396, 2023.
- B-cos networks: Alignment is all we need for interpretability. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 10329–10338, 2022.
- This looks like that: deep learning for interpretable image recognition. Advances in neural information processing systems, 32, 2019.
- Transmed: Transformers advance multi-modal medical image classification. Diagnostics, 11(8):1384, 2021.
- An image is worth 16x16 words: Transformers for image recognition at scale. In International Conference on Learning Representations, 2021. URL https://openreview.net/forum?id=YicbFdNTTy.
- Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770–778, 2016.
- Adversarial unsupervised domain adaptation with conditional and label shift: Infer, align and iterate. In Proceedings of the IEEE/CVF international conference on computer vision, pp. 10367–10376, 2021a.
- Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF international conference on computer vision, pp. 10012–10022, 2021b.
- Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023.
- Moment matching for multi-source domain adaptation. In Proceedings of the IEEE/CVF international conference on computer vision, pp. 1406–1415, 2019.
- Adapting visual category models to new domains. In Computer Vision–ECCV 2010: 11th European Conference on Computer Vision, Heraklion, Crete, Greece, September 5-11, 2010, Proceedings, Part IV 11, pp. 213–226. Springer, 2010.
- Maximum classifier discrepancy for unsupervised domain adaptation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 3723–3732, 2018.
- Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision, pp. 618–626, 2017.
- Deep inside convolutional networks: visualising image classification models and saliency maps. In Proceedings of the International Conference on Learning Representations (ICLR). ICLR, 2014.
- Revisiting weakly supervised pre-training of visual perception models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 804–814, 2022.
- Prototypical networks for few-shot learning. Advances in neural information processing systems, 30, 2017.
- Adversarial discriminative domain adaptation. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 7167–7176, 2017.
- Attention is all you need. Advances in neural information processing systems, 30, 2017.
- Central moment discrepancy (cmd) for domain-invariant representation learning. arXiv preprint arXiv:1702.08811, 2017.
- Delving deep into the generalization of vision transformers under distribution shifts. In Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition, pp. 7277–7286, 2022.
- Multiple source domain adaptation with adversarial learning. 2018.
Sponsor
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.
