Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Domain Adaptation With Similarity Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace the fully connected classifier with prototype-based similarity learning | RevGrad domain adaptation on VisDA · VisDA-val synthetic to real | per-class average accuracy (%): 58.62 -> 69.58 | Helped |
| use prototype-based similarity classification | domain adaptation on Office-31 · Office-31 average | accuracy (%): 84.6 -> 86.2 | Helped |
| use prototype-based similarity classification | domain adaptation on MNIST to MNIST-M · MNIST to MNIST-M | per-class average accuracy (%): 98.2 -> 90.5 | Hurt |
| share the weights of the input and prototype networks (ablation) | SimNet · VisDA-val | average per-class accuracy (%): 69.6 -> 57.5 | Hurt |
| add the prototype orthogonality regularizer (ablation) | SimNet · VisDA-val | average per-class accuracy (%): 66.1 -> 69.6 | Helped |
| use ResNet-152 instead of ResNet-50 (ablation) | SimNet · VisDA-val | average per-class accuracy (%): 69.6 -> 72.9 | Helped |
| use the similarity-based classifier instead of a fully connected classifier | source-only classification on VisDA · VisDA-val | average per-class accuracy (%): 46.0 -> 49.5 | Helped |
Closest papers by meaning
- Co-regularized Alignment for Unsupervised Domain Adaptation (NeurIPS 2018 · similarity 0.92)
- Learning Transferrable Representations for Unsupervised Domain Adaptation (NeurIPS 2016 · similarity 0.92)
- Maximum Classifier Discrepancy for Unsupervised Domain Adaptation (CVPR 2018 · similarity 0.91)
- Unsupervised Domain Adaptation Using Feature-Whitening and Consensus Loss (CVPR 2019 · similarity 0.91)
- Unsupervised Domain Adaptation for Distance Metric Learning (ICLR 2019 · similarity 0.91)
- Low Category Uncertainty and High Training Potential Instance Learning for Unsupervised Domain Adaptation (AAAI 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.