Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
MetaGAN: An Adversarial Approach to Few-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 task-conditioned adversarial generator and fake class | Relation Net · Mini-Imagenet 5-way 1-shot | accuracy: 50.44 -> 52.71 | Helped |
| add MetaGAN | Relation Net · Mini-Imagenet 5-way 5-shot | accuracy: 65.32 -> 68.63 | Helped |
| add MetaGAN | MAML (1 step, first order) · Mini-Imagenet 5-way 1-shot | accuracy: 43.64 -> 46.13 | Helped |
| add MetaGAN | MAML · Omniglot 5-way 5-shot | accuracy: 99.9 -> 99.7 | No clear effect |
| add MetaGAN | Relation Net · Omniglot 20-way 1-shot | accuracy: 97.6 -> 97.64 | No clear effect |
| add MetaGAN using unlabeled samples | Relation Net · Mini-Imagenet 1-shot sample-level semi-supervised | accuracy: 44.24 -> 50.35 | Helped |
| add MetaGAN learning from unlabeled tasks | Relation Net · Mini-Imagenet 1-shot task-level semi-supervised | accuracy: 43.87 -> 47.43 | Helped |
Closest papers by meaning
- Adversarially Robust Few-Shot Learning: A Meta-Learning Approach (NeurIPS 2020 · similarity 0.90)
- Dense Classification and Implanting for Few-Shot Learning (CVPR 2019 · similarity 0.90)
- A Baseline for Few-Shot Image Classification (ICLR 2020 · similarity 0.89)
- Meta-Learning With Differentiable Convex Optimization (CVPR 2019 · similarity 0.89)
- Unraveling Meta-Learning: Understanding Feature Representations for Few-Shot Tasks (ICML 2020 · similarity 0.89)
- Pushing the Limit of Fine-Tuning for Few-Shot Learning: Where Feature Reusing Meets Cross-Scale Attention (AAAI 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.