Research map / Label-efficient and robust learning / Recognition and long-tail
Learning Stable Classifiers by Transferring Unstable Features
Yujia Bao, Shiyu Chang, Dr.Regina Barzilay
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 |
|---|---|---|---|
| Transfer unstable-feature information from source environments, cluster target examples by that representation, and train with worst-cluster risk | TOFU · 12 synthetic transfer settings | accuracy: 22.9% | Improved |
| Transfer unstable-feature information from source environments and optimize worst-cluster risk | TOFU · Synthetic transfer settings | accuracy: 0.30% absolute difference | Improved |
| Transfer unstable-feature information from source environments and optimize worst-cluster risk | TOFU · CelebA, 38 latent attributes | worst-group accuracy: 18.06% | Improved |
| Directly reuse or fine-tune a source-task feature extractor instead of transferring an unstable-feature representation | Target-task transfer methods · Synthetic transfer settings | accuracy | Hurt |
Closest papers by meaning
- Stable Classification (JMLR 2022 · similarity 0.89)
- Spuriosity Didn’t Kill the Classifier: Using Invariant Predictions to Harness Spurious Features (NeurIPS 2023 · similarity 0.87)
- Building a stable classifier with the inflated argmax (NeurIPS 2024 · similarity 0.87)
- Mitigating Biases in Blackbox Feature Extractors for Image Classification Tasks (NeurIPS 2024 · similarity 0.85)
- Classifier-to-Bias: Toward Unsupervised Automatic Bias Detection for Visual Classifiers (CVPR 2025 · similarity 0.84)
- Classes Are Not Equal: An Empirical Study on Image Recognition Fairness (CVPR 2024 · similarity 0.84)
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.
