Research map / Label-efficient and robust learning / Domain adaptation and generalization
Anti-causal domain generalization: Leveraging unlabeled data
Sorawit Saengkyongam, Juan L. Gamella, Andrew Miller, Jonas Peters, Nicolai Meinshausen, Christina Heinze-Deml
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add MIR (mean-based invariant regularization using unlabeled environments) vs Pooled-Ridge | linear regression · Light Tunnel (leave-one-environment-out) | RMSE | Improved |
| add VIR (variance-based invariant regularization) to a linear head on pretrained embeddings | Pooled-Ridge · VitalDB stroke volume (subject-centered) | nMSE | Improved |
| use VIR instead of Pooled-Ridge when few environments are labeled (ablation) | VIR · VitalDB (20 labeled subjects) | average nMSE | Improved |
| increase number of unlabeled subjects from 0 to 80 (20 labeled) (ablation) | VIR · VitalDB | nMSE | Improved |
| VIR vs Pooled-Ridge on tracking quality | VIR · VitalDB | Spearman correlation | No clear effect |
| tune gamma on the test environment (oracle) instead of cross-validation (ablation) | MIR / VIR · Light Tunnel and VitalDB | RMSE / nMSE | Improved |
| reverse prediction direction (violate anti-causal assumption) (ablation) | MIR · Light Tunnel reversed (predict ir_1 from RGB) | RMSE | No clear effect |
| remove subject-centering (environment affects outcome) (ablation) | MIR-VIR · VitalDB without subject-centering | average nMSE | Mixed |
Closest papers by meaning
- On Causality in Domain Adaptation and Semi-Supervised Learning: an Information-Theoretic Analysis for Parametric Models (JMLR 2024 · similarity 0.90)
- Invariant Models for Causal Transfer Learning (JMLR 2018 · similarity 0.90)
- Do causal predictors generalize better to new domains? (NeurIPS 2024 · similarity 0.89)
- Invariant and Transportable Representations for Anti-Causal Domain Shifts (NeurIPS 2022 · similarity 0.89)
- Domain Generalization using Causal Matching (ICML 2021 · similarity 0.89)
- Understanding Domain Generalization: A Noise Robustness Perspective (ICLR 2024 · similarity 0.88)
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.
