Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
A Bit More Bayesian: Domain-Invariant Learning with Uncertainty
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 Bayesian probabilistic domain-invariant learning in the classifier and feature extractor (ablation) | ResNet-18 · PACS | mean accuracy: 80.28 { $\pm$0.42} -> 85.45 { $\pm$0.24} | Helped |
| Use Bayesian domain-invariant learning | ResNet-18 · Office-Home | mean accuracy: 65.63 -> 66.42 { $\pm$0.18} | Helped |
| Use Bayesian domain-invariant learning | ResNet-18 · Rotated MNIST out-of-distribution | accuracy: 94.7 -> 96.5 { $\pm$0.08} | Helped |
| Use Bayesian domain-invariant learning | ResNet-18 · Fashion-MNIST out-of-distribution | accuracy: 78.0 -> 83.5 { $\pm$0.63} | Helped |
| Add Bayesian inference to the classifier (ablation) | PACS ResNet-18 baseline · PACS | mean accuracy: 80.28 { $\pm$0.42} -> 81.93 { $\pm$0.22} | Helped |
| Add deterministic domain-invariant learning to the classifier (ablation) | PACS ResNet-18 baseline · PACS | mean accuracy: 80.28 { $\pm$0.42} -> 82.46 { $\pm$0.67} | Helped |
| Replace standard Gaussian prior with scale mixture prior (ablation) | Bayesian domain-invariant learning · PACS | mean accuracy: 83.61 { $\pm$0.12} -> 85.45 { $\pm$0.24} | Helped |
| Add a Bayesian feature layer without domain invariance (ablation) | Bayesian domain-invariant learning · PACS | mean accuracy: 85.45 { $\pm$0.24} -> 85.51 { $\pm$0.13} | No clear effect |
| Add a Bayesian feature layer with domain-invariant learning (ablation) | Bayesian domain-invariant learning · PACS | mean accuracy: 85.45 { $\pm$0.24} -> 84.96 { $\pm$0.34} | Hurt |
| Set mixture-prior parameter π to 0.5 (ablation) | Bayesian domain-invariant learning · PACS cartoon | accuracy | Helped |
Closest papers by meaning
- Robust Domain Generalization under Divergent Marginal and Conditional Distributions (NeurIPS 2026 · similarity 0.89)
- Domain Generalization via Conditional Invariant Representations (AAAI 2018 · similarity 0.89)
- Learning Invariant Representations and Risks for Semi-Supervised Domain Adaptation (CVPR 2021 · similarity 0.89)
- Learning Informative Invariant Representations via Hierarchical Latent Decomposition (NeurIPS 2026 · similarity 0.89)
- Domain Adaptation with Invariant Representation Learning: What Transformations to Learn? (NeurIPS 2021 · similarity 0.88)
- Failure Modes of Domain Generalization Algorithms (CVPR 2022 · 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.