Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Adaptation by Mixture of Alignments of Second- or Higher-Order Scatter Tensors
Check 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 |
|---|---|---|---|
| align class-specific second-order scatters | AlexNet source and target streams · Office A→W | accuracy: 82.5±0.9 -> 84.5±1.7 | Helped |
| align class-specific second-order scatters | VGG source and target streams · Office A→D | accuracy: 89.83±1.4 -> 92.26±1.1 | Helped |
| add class-specific weights to second-order alignment (ablation) | VGG So · Office A→W | accuracy: 89.50±1.8 -> 90.24±1.3 | Helped |
| use third-order alignment with class-specific weights instead of unweighted second-order alignment (ablation) | VGG · Office A→W | accuracy: 89.50±1.8 -> 90.55±1.1 | Helped |
| combine second- and third-order alignment with class-specific weights (ablation) | VGG · Office A→W | accuracy: 89.50±1.8 -> 90.75±1.2 | Helped |
| add fourth-order alignment to weighted second- and third-order alignment (ablation) | VGG · Office A→W | accuracy: 90.75±1.2 -> 90.92±1.3 | Helped |
| add fourth-order alignment to weighted second- and third-order alignment (ablation) | VGG · Office A→D | accuracy: 92.52±1.2 -> 92.73±1.1 | Helped |
| use second-order alignment | AlexNet · Pascal VOC2007→TU Berlin | accuracy: 62.66 -> 63.40 | Helped |
| increase target training examples | So domain adaptation · RGB-D-Caltech256 | accuracy | Mixed |
| align second-order marginal statistics rather than class-specific scatters (ablation) | AlexNet · Office A→W | accuracy: 82.6% vs 84.5% | Hurt |
| align second-order marginal statistics rather than class-specific scatters (ablation) | AlexNet · Office D→W | accuracy: 94.6% vs 95.5% | Hurt |
| align second-order marginal statistics rather than class-specific scatters (ablation) | AlexNet · Office W→D | accuracy: 95.9% vs 97.5% | Hurt |
Closest papers by meaning
- Adapted Deep Embeddings: A Synthesis of Methods for k-Shot Inductive Transfer Learning (NeurIPS 2018 · similarity 0.88)
- Unsupervised Domain Adaptation with Residual Transfer Networks (NeurIPS 2016 · similarity 0.88)
- Boosting Domain Adaptation by Discovering Latent Domains (CVPR 2018 · similarity 0.88)
- HiGDA: Hierarchical Graph of Nodes to Learn Local-to-Global Topology for Semi-Supervised Domain Adaptation (AAAI 2025 · similarity 0.88)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · 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.