Research map / Label-efficient and robust learning / Domain adaptation and generalization
Towards Accurate and Robust Domain Adaptation Under Multiple Noisy Environments
Zhongyi Han, Xian-Jin Gui, Haoliang Sun, Yilong Yin, Shuo Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 offline curriculum learning (ablation) | RDA · Office-31 (40% mixed corruption) | accuracy (%): 81.8 -> 86.4 | Improved |
| add proxy margin discrepancy (ablation) | RDA · Office-31 (40% mixed corruption) | accuracy (%): 85.2 -> 86.4 | Improved |
| use class-aware averaged-loss offline curriculum instead of online curriculum (ablation) | RDA · Office-31 (40% label corruption) | accuracy (%): 73.3 -> 83.6 | Improved |
| use averaged loss and class priors in offline curriculum rather than plain offline (ablation) | RDA · Office-31 (40% label corruption) | accuracy (%): 75.9 -> 83.6 | Improved |
| reserve feature-noise examples instead of deleting them (ablation) | RDA · Office-31 (40% feature corruption) | accuracy (%): 85.4 -> 87.2 | Improved |
| treat label-noise and feature-noise separately rather than indiscriminately (vs TCL) | RDA · Office-31 (40% feature corruption) | accuracy (%): 79.9 -> 87.2 | Improved |
| increase feature noise from 0% to 80% | RDA · Office-31 A->W | accuracy (%): 95.9 -> 80.4 | Hurt |
| increase feature noise from 0% to 80% | TCL · Office-31 A->W | accuracy (%): 87.5 -> 31.2 | Hurt |
| increase feature noise from 0% to 80% | DANN · Office-31 A->W | accuracy (%): 85.5 -> 53.2 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MentorNet — MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels (ICML 2018) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Label-Noise Robust Domain Adaptation (ICML 2020 · similarity 0.91)
- De-Confusing Pseudo-Labels in Source-Free Domain Adaptation (ECCV 2024 · similarity 0.91)
- A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation (ECCV 2020 · similarity 0.91)
- Bridging Theory and Algorithm for Domain Adaptation (ICML 2019 · similarity 0.90)
- When Source-Free Domain Adaptation Meets Learning with Noisy Labels (ICLR 2023 · similarity 0.90)
- Confusing Pair Correction Based on Category Prototype for Domain Adaptation under Noisy Environments (AAAI 2024 · similarity 0.90)
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.
