Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Class Normalization for (Continual)? Generalized Zero-Shot Learning
Read the paperCheck 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 |
|---|---|---|---|
| Add class normalization to a 3-layer MLP (ablation) | Embedding-based ZSL · SUN, CUB, AwA1, and AwA2 | H: SUN 35.3 -> 43.1; CUB 46.7 -> 50.3; AwA1 62.8 -> 67.8; AwA2 62.1 -> 67.6 | Mixed |
| Add class normalization to EWC-online (ablation) | Continual zero-shot learning · CUB and SUN | mH: CUB 18.0 -> 23.3; SUN 9.6 -> 14.3 | Helped |
| Add class normalization to MAS-online (ablation) | Continual zero-shot learning · CUB and SUN | mH: CUB 17.7 -> 23.8; SUN 9.4 -> 14.2 | Helped |
| Add class normalization to A-GEM (ablation) | Continual zero-shot learning · CUB and SUN | mH: CUB 17.3 -> 23.8; SUN 9.6 -> 14.2 | Helped |
| Add class normalization to Sequential (ablation) | Continual zero-shot learning · CUB and SUN | mH: CUB 17.2 -> 23.0; SUN 9.3 -> 14.0 | Helped |
| Add class normalization to Multi-task (ablation) | Continual zero-shot learning · CUB and SUN | mH: CUB 24.3 -> 30.0; SUN 12.5 -> 14.8 | Helped |
| Use attributes normalization alone in a deep attribute embedder (ablation) | Embedding-based ZSL · Deep attribute embedder | representation variance stability | Hurt |
Closest papers by meaning
- Adaptive Confidence Smoothing for Generalized Zero-Shot Learning (CVPR 2019 · similarity 0.89)
- Unseen Classes at a Later Time? No Problem (CVPR 2022 · similarity 0.88)
- Class Distribution Shifts in Zero-Shot Learning: Learning Robust Representations (NeurIPS 2024 · similarity 0.87)
- Data-Free Generalized Zero-Shot Learning (AAAI 2024 · similarity 0.87)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.86)
- Data Distribution Distilled Generative Model for Generalized Zero-Shot Recognition (AAAI 2024 · 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.