Research map / Label-efficient and robust learning / OOD and anomaly detection
Invariance Principle Meets Information Bottleneck for Out-of-Distribution 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 |
|---|---|---|---|
| add IB (variance) penalty | ERM · CS-CMNIST | accuracy: 60.27 -> 71.80 | Helped |
| add IB penalty | ERM · AC-CMNIST | accuracy: 16.84 -> 50.24 | Helped |
| add IB penalty | ERM · COCO | accuracy: 22.70 -> 31.66 | Helped |
| add IB penalty | ERM · Terra Incognita | accuracy: 49.80 -> 56.40 | Helped |
| add IB penalty | IRM · CS-CMNIST | accuracy: 61.49 -> 71.79 | Helped |
| add IB penalty | IRM · COCO | accuracy: 18.47 -> 25.10 | Helped |
| add IB penalty | IRM · AC-CMNIST | accuracy: 66.98 -> 67.67 | No clear effect |
| add IB penalty | IRM · Terra Incognita | accuracy: 54.60 -> 54.10 | No clear effect |
| add IB penalty | IRM · Linear Example2 (6 envs) | classification error: 0.46 -> 0.43 | No clear effect |
| add IRM penalty | ERM · AC-CMNIST | accuracy: 16.84 -> 66.98 | Helped |
| add IRM penalty to ERM | ERM · CS-CMNIST | accuracy: 60.27 -> 61.49 | No clear effect |
| add IB penalty | ERM · Linear Example1 (3 envs) | MSE: 13.36 -> 12.96 | No clear effect |
Closest papers by meaning
- Diagnosing and Rectifying Fake OOD Invariance: A Restructured Causal Approach (AAAI 2024 · similarity 0.89)
- On the Out-of-Distribution Generalization of Self-Supervised Learning (ICML 2025 · similarity 0.87)
- Dissecting the Failure of Invariant Learning on Graphs (NeurIPS 2024 · similarity 0.87)
- Aggregation Hides Out-of-Distribution Generalization Failures from Spurious Correlations (NeurIPS 2025 · similarity 0.87)
- Empowering Graph Invariance Learning with Deep Spurious Infomax (ICML 2024 · similarity 0.87)
- Out-of-Variable Generalisation for Discriminative Models (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.