Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning to Adapt to Evolving Domains
Hong Liu, Mingsheng Long, Jianmin Wang, Yu Wang
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 meta-representation (meta-objective for evolving domains) (ablation) | Source/baseline adapter · Rotated MNIST | avg accuracy: 26.98 -> 33.57 (EAML rep vs Source Only) | Improved |
| add meta-adapter on top of meta-representation (ablation) | EAML (rep) · Rotated MNIST | avg accuracy: 33.57 -> 35.23 | Improved |
| add meta-representation on top of meta-adapter (ablation) | EAML (ada) · Rotated MNIST | avg accuracy: 33.19 -> 35.23 | Improved |
| add meta-adapter to EAML (rep) at last target (174 deg) (ablation) | EAML (rep) · Rotated MNIST 174 deg | accuracy: 43.30 -> 42.27 | Hurt |
| use EAML instead of merging target data and adapting offline with JAN/CDAN | domain adaptation · Rotated MNIST | avg accuracy: 32.87 -> 35.23 | Improved |
| use meta-learning (EAML) instead of generative replay oracle (DANN+Replay) | evolving DA · Rotated MNIST | avg accuracy: 34.28 -> 35.23 | Improved |
| add EWC regularization to DANN | DANN · Rotated MNIST | avg accuracy: 29.81 -> 32.27 | Improved |
| vary lambda and meta-adapter learning rate (ablation) | EAML · Caltran | accuracy | No clear effect |
Closest papers by meaning
- Online Meta-Learning for Multi-Source and Semi-Supervised Domain Adaptation (ECCV 2020 · similarity 0.89)
- Meta-DMoE: Adapting to Domain Shift by Meta-Distillation from Mixture-of-Experts (NeurIPS 2022 · similarity 0.89)
- Continuous Adaptation via Meta-Learning in Nonstationary and Competitive Environments (ICLR 2018 · similarity 0.88)
- A Comprehensive Survey on Source-Free Domain Adaptation (TPAMI 2024 · similarity 0.88)
- Diffusion-Driven Progressive Target Manipulation for Source-Free Domain Adaptation (NeurIPS 2025 · similarity 0.87)
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · 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.
