Research map / Label-efficient and robust learning / Domain adaptation and generalization
Meta-Causal Learning for Single Domain Generalization
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 |
|---|---|---|---|
| Use meta-causal learning with simulate-analyze-reduce | Single-domain generalization · Digits (MNIST source; four target domains) | average accuracy (%): 74.77 -> 78.82 | Helped |
| Use meta-causal learning with simulate-analyze-reduce | Single-domain generalization · CIFAR10-C (levels 1–5) | average accuracy (%): 85.55 -> 89.15 | Helped |
| Use meta-causal learning with simulate-analyze-reduce | Single-domain generalization · PACS | average accuracy (%): 68.10 -> 69.86 | Helped |
| Use meta-causal learning | PACS leave-one-domain-out · PACS | average accuracy (%): 86.32 -> 87.00 | Helped |
| Add domain transformation to the base module (ablation) | Meta-causal learning · PACS | average accuracy (%): 54.97 -> 67.32 | Helped |
| Add domain alignment without counterfactual inference (ablation) | Meta-causal learning with domain transformation · PACS | average accuracy (%): 67.32 -> 63.39 | Hurt |
| Add counterfactual inference to domain transformation and alignment (ablation) | Meta-causal learning · PACS | average accuracy (%): 63.39 -> 69.86 | Helped |
| Use photometric factors only instead of geometric factors only (ablation) | Meta-causal learning · PACS | average accuracy (%): 60.12 -> 62.49 | Helped |
| Use both photometric and geometric factors instead of photometric factors only (ablation) | Meta-causal learning · PACS | average accuracy (%): 62.49 -> 69.86 | Helped |
Closest papers by meaning
- Learning to Generalize: Meta-Learning for Domain Generalization (AAAI 2018 · similarity 0.91)
- Learning to Generalize across Domains on Single Test Samples (ICLR 2022 · similarity 0.91)
- Causality Inspired Representation Learning for Domain Generalization (CVPR 2022 · similarity 0.89)
- Domain Generalization using Causal Matching (ICML 2021 · similarity 0.89)
- Enhancing Evolving Domain Generalization through Dynamic Latent Representations (AAAI 2024 · similarity 0.87)
- Adapting to Distribution Shift by Visual Domain Prompt Generation (ICLR 2024 · similarity 0.87)
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.