Research map / Label-efficient and robust learning / Domain adaptation and generalization
Progressive Domain Expansion Network for Single Domain Generalization
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 |
|---|---|---|---|
| Progressively generate domains and train with them using PDEN | Single-domain generalization · Digits (MNIST source; target Avg.) | accuracy: 59.49 -> 74.77 | Helped |
| Use PDEN instead of RandAugment | CIFAR10-C · Level1 | accuracy: 91.74 -> 90.62 | Hurt |
| Use PDEN instead of RandAugment | CIFAR10-C · Level2 | accuracy: 88.89 -> 88.91 | Helped |
| Use PDEN instead of RandAugment | CIFAR10-C · Level3 | accuracy: 85.82 -> 87.03 | Helped |
| Use PDEN instead of RandAugment | CIFAR10-C · Level4 | accuracy: 81.03 -> 83.71 | Helped |
| Use PDEN instead of RandAugment | CIFAR10-C · Level5 | accuracy: 74.93 -> 77.47 | Helped |
| Increase the number of progressively generated domains K (ablation) | PDEN · Digits | accuracy | Helped |
| Increase the adversarial-loss weight w_adv (ablation) | PDEN · Digits unseen target domains | accuracy | Helped |
| Increase the cycle-consistency weight w_cyc (ablation) | PDEN · MNIST-M, SVHN, SYNDIGIT | accuracy | Helped |
| Increase the cycle-consistency weight w_cyc (ablation) | PDEN · USPS | accuracy | No clear effect |
| Increase the diversity-loss weight w_div (ablation) | PDEN · Digits unseen target domains | accuracy | Helped |
Closest papers by meaning
- Domain Invariant Representation Learning with Domain Density Transformations (NeurIPS 2021 · similarity 0.89)
- Towards Unsupervised Domain Generalization (CVPR 2022 · similarity 0.88)
- Adaptive Methods for Real-World Domain Generalization (CVPR 2021 · similarity 0.88)
- Gradient-Guided Annealing for Domain Generalization (CVPR 2025 · similarity 0.87)
- PracticalDG: Perturbation Distillation on Vision-Language Models for Hybrid Domain Generalization (CVPR 2024 · similarity 0.86)
- Seeking Consistent Flat Minima for Better Domain Generalization via Refining Loss Landscapes (CVPR 2025 · similarity 0.86)
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.