Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Few-Shot Open-Set Recognition Using Meta-Learning
Check 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 GaussianE to the Euclidean-distance baseline (ablation) | PEELER · CIFAR10 (ConvNet) | AUROC (%): 74.62 -> 75.65 | Helped |
| Add OpLoss to the GaussianE model (ablation) | PEELER · CIFAR10 (ConvNet) | AUROC (%): 75.65 -> 77.22 | Helped |
| Add GaussianE to the Euclidean-distance baseline (ablation) | PEELER · CIFAR10 (ResNet18) | AUROC (%): 82.94 -> 83.12 | Helped |
| Add OpLoss to the GaussianE model (ablation) | PEELER · CIFAR10 (ResNet18) | AUROC (%): 83.12 -> 83.99 | Helped |
| Add GaussianE to the Euclidean-distance baseline (ablation) | PEELER · miniImageNet (large-scale) | AUROC (%): 80.59 -> 81.06 | Helped |
| Add OpLoss to the GaussianE model (ablation) | PEELER · miniImageNet (large-scale) | AUROC (%): 81.06 -> 82.12 | Helped |
| Add OpLoss to the Euclidean-distance baseline (ablation) | PEELER · miniImageNet 5-way 1-shot | AUROC (%): 58.94±0.60 -> 60.94±0.61 | Helped |
| Add GaussianE without OpLoss (ablation) | PEELER · miniImageNet 5-way 1-shot | AUROC (%): 58.94±0.60 -> 58.66±0.60 | No clear effect |
| Add OpLoss to the Euclidean-distance baseline (ablation) | PEELER · miniImageNet 5-way 5-shot | AUROC (%): 66.00±0.67 -> 67.92±0.68 | Helped |
| Add GaussianE without OpLoss (ablation) | PEELER · miniImageNet 5-way 5-shot | AUROC (%): 66.00±0.67 -> 66.50±0.67 | No clear effect |
| Use more classes in the few-shot recognition task (5-way to 10-way) (ablation) | PEELER · miniImageNet 1-shot open-set recognition | AUROC (%): 61.66±0.62 -> 74.97±0.74 | Helped |
| Vary the number of unseen classes sampled per training episode while keeping total open-set samples fixed (ablation) | PEELER · Open-set recognition | AUROC | No clear effect |
Closest papers by meaning
- Open-Set Recognition: A Good Closed-Set Classifier is All You Need (ICLR 2022 · similarity 0.90)
- Meta Evidential Transformer for Few-Shot Open-Set Recognition (ICML 2024 · similarity 0.89)
- Few-Shot Open-Set Recognition by Transformation Consistency (CVPR 2021 · similarity 0.89)
- Open-Set Likelihood Maximization for Few-Shot Learning (CVPR 2023 · similarity 0.88)
- Generative-Discriminative Feature Representations for Open-Set Recognition (CVPR 2020 · similarity 0.87)
- Frequency Shuffling and Enhancement for Open Set Recognition (AAAI 2024 · similarity 0.85)
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.