Research map / Label-efficient and robust learning / Domain adaptation and generalization
On Learning Domain-Invariant Representations for Transfer Learning with Multiple Sources
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 just-right compressed DI representations instead of general DI representations | Colored MNIST DG · Colored MNIST, close target domain (θ=0.7) | target accuracy | Helped |
| Use general DI representations (λ=0) instead of compressed DI representations | Colored MNIST DG · Colored MNIST, far target domain (θ=0.05) | target accuracy | Helped |
| Increase compression strength beyond the just-right level (ablation) | Colored MNIST DG · Colored MNIST, close target domain (θ=0.7) | target accuracy | Hurt |
| Increase source compression from general DI representations to an appropriate level (ablation) | Colored MNIST DG · Colored MNIST, source validation | source validation accuracy | Helped |
| Increase source compression beyond the appropriate level (ablation) | Colored MNIST DG · Colored MNIST, source validation | source validation accuracy | Hurt |
| Use a far rather than a close target domain | Colored MNIST DG · Colored MNIST, target domains θ=0.05 and θ=0.7 | target accuracy | Hurt |
Closest papers by meaning
- Domain Adaptation with Invariant Representation Learning: What Transformations to Learn? (NeurIPS 2021 · similarity 0.91)
- Learning to Transfer with von Neumann Conditional Divergence (AAAI 2022 · similarity 0.90)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.90)
- On Learning Invariant Representations for Domain Adaptation (ICML 2019 · similarity 0.89)
- Learning Invariant Representations and Risks for Semi-Supervised Domain Adaptation (CVPR 2021 · similarity 0.89)
- Wasserstein Transfer Learning (NeurIPS 2025 · 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.