Research map / Label-efficient and robust learning / Domain adaptation and generalization
Meta Convolutional Neural Networks for Single Domain Generalization
Check 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 baseline with MetaCNN using bag-of-meta-feature composition | Digits models trained on MNIST · Digits target Avg | accuracy: 74.46 -> 78.76 | Helped |
| replace the baseline with MetaCNN using bag-of-meta-feature composition | WideResNet models trained on CIFAR-10 · CIFAR-10-C severity 5 Avg | accuracy: 72.61 -> 78.45 | Helped |
| replace the baseline with MetaCNN using bag-of-meta-feature composition | AlexNet models trained on PACS photo · PACS target Avg | accuracy: 55.24 -> 57.17 | Helped |
| add local feature addressing to select related meta features (ablation) | MetaCNN · Digits target Avg | accuracy: 74.28 -> 78.76 | Helped |
| increase the number of BoMF blocks from one to two (ablation) | MetaCNN · Digits target Avg | accuracy: 77.21 -> 78.76 | Helped |
| increase the number of BoMF blocks from two to three (ablation) | MetaCNN · Digits target Avg | accuracy: 78.76 -> 79.34 | Helped |
| increase the number of BoMF blocks from three to four (ablation) | MetaCNN · Digits target Avg | accuracy: 79.34 -> 79.24 | Hurt |
| add BoMF composition | AlexNet trained on PACS photo · PACS art painting | accuracy: 56.26 -> 54.05 | Hurt |
| add BoMF composition | AlexNet trained on PACS photo · PACS cartoon and sketch | accuracy: 51.04 -> 53.58; 58.42 -> 63.88 | Helped |
Closest papers by meaning
- Open Domain Generalization with Domain-Augmented Meta-Learning (CVPR 2021 · similarity 0.90)
- MetaReg: Towards Domain Generalization using Meta-Regularization (NeurIPS 2018 · similarity 0.88)
- CLIPCEIL: Domain Generalization through CLIP via Channel rEfinement and Image-text aLignment (NeurIPS 2024 · similarity 0.87)
- Unsupervised Meta-Learning via In-Context Learning (ICLR 2025 · similarity 0.86)
- Leveraging Vision-Language Models for Improving Domain Generalization in Image Classification (CVPR 2024 · similarity 0.86)
- Domain-Controlled Prompt Learning (AAAI 2024 · similarity 0.86)
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.