Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Co-Representation Network for Generalized Zero-Shot Learning
Fei Zhang, Guangming Shi
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 |
|---|---|---|---|
| replace two-layer MLP projection with cooperation module (parallel single-layer experts) (ablation) | RN · AwA2 | unseen accuracy Au: 30.0 -> 52.6 | Improved |
| replace two-layer MLP with cooperation module (ablation) | RN · AwA1 | unseen accuracy Au: 31.4 -> 58.1 | Improved |
| replace two-layer MLP with cooperation module (ablation) | RN · CUB | unseen accuracy Au: 38.1 -> 45.5 | Improved |
| replace two-layer MLP with cooperation module (ablation) | RN · AwA2 | seen accuracy As: 93.4 -> 78.8 | Hurt |
| replace two-layer MLP with cooperation module (ablation) | RN · AwA1 | harmonic mean H: 46.7 -> 65.4 | Improved |
| replace two-layer MLP with cooperation module (ablation) | RN · AwA2 | average LRD: 0.756 -> 0.956 | Improved |
| replace two-layer MLP with cooperation module (ablation) | RN · CUB | average LRD: 0.831 -> 0.843 | Improved |
| replace two-layer MLP with cooperation module (ablation) | RN · AwA1 | average LRD: 0.711 -> 0.835 | Improved |
| embedding model CRnet vs generative model f-CLSWGAN | CRnet · SUN | harmonic mean H: 35.3 vs 39.4 | Hurt |
| CRnet vs generative SE-GZSL | CRnet · AwA2 | seen accuracy As: 68.1 -> 78.8 | Improved |
| slightly increase number of experts K (ablation) | CRnet · all datasets | accuracy | No clear effect |
| set beta regularization to 0 instead of large value on fine-grained datasets (ablation) | CRnet · CUB, SUN | fitting | Improved |
Closest papers by meaning
- Generalized Zero-Shot Learning with Deep Calibration Network (NeurIPS 2018 · similarity 0.90)
- An Empirical Study and Analysis of Generalized Zero-Shot Learning for Object Recognition in the Wild (ECCV 2016 · similarity 0.89)
- Rethinking Zero-Shot Learning: A Conditional Visual Classification Perspective (ICCV 2019 · similarity 0.88)
- Zero-Shot Recognition via Semantic Embeddings and Knowledge Graphs (CVPR 2018 · similarity 0.88)
- Predicting Visual Exemplars of Unseen Classes for Zero-Shot Learning (ICCV 2017 · similarity 0.88)
- Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names (CVPR 2024 · similarity 0.88)
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.
