Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Meta-Learning with Adaptive Hyperparameters
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 |
|---|---|---|---|
| Use ALFA adaptive inner-loop updates instead of standard MAML updates | MAML with ResNet12 · miniImageNet 5-way 5-shot | accuracy: 69.76 \pm 0.46\% -> 77.96 \pm 0.41\% | Helped |
| Add ALFA adaptive updates to MAML + L2F | MAML + L2F with ResNet12 · miniImageNet 5-way 5-shot | accuracy: 77.04 \pm 0.42\% -> 77.42 \pm 0.42\% | No clear effect |
| Use ALFA adaptive inner-loop updates instead of standard MAML updates | MAML with ResNet12 · tieredImageNet 5-way 5-shot | accuracy: 71.24 \pm 0.43\% -> 82.48 \pm 0.38\% | Helped |
| Add ALFA adaptive updates to MAML + L2F | MAML + L2F with ResNet12 · tieredImageNet 5-way 5-shot | accuracy: 81.13 \pm 0.39\% -> 81.77 \pm 0.39\% | Helped |
| Use ALFA adaptive inner-loop updates instead of standard MAML updates | MAML with ResNet12 · FC100 5-way 5-shot | accuracy: 52.63 \pm 0.50\% -> 55.82 \pm 0.50\% | Helped |
| Add ALFA adaptive updates to MAML + L2F | MAML + L2F with ResNet12 · FC100 5-way 5-shot | accuracy: 54.68 \pm 0.50\% -> 55.23 \pm 0.50\% | Helped |
| Use ALFA adaptive inner-loop updates instead of standard MAML updates | MAML with ResNet12 · CIFAR-FS 5-way 5-shot | accuracy: 76.38 \pm 0.42\% -> 83.62 \pm 0.37\% | Helped |
| Add ALFA adaptive updates to MAML + L2F | MAML + L2F with ResNet12 · CIFAR-FS 5-way 5-shot | accuracy: 82.79 \pm 0.38\% -> 82.98 \pm 0.38\% | No clear effect |
| Use ALFA adaptive inner-loop updates instead of standard MAML updates | MAML with 4-CONV · miniImageNet → Omniglot 5-way 5-shot | accuracy: 85.68 \pm 0.35\% -> 93.11 \pm 0.23\% | Helped |
| Add ALFA adaptive updates to MAML + L2F | MAML + L2F with 4-CONV · miniImageNet → Omniglot 5-way 5-shot | accuracy: 94.96 \pm 0.22\% -> 94.10 \pm 0.24\% | Hurt |
| Use ALFA adaptive inner-loop updates instead of standard MAML updates | MAML with 4-CONV · miniImageNet → FC100 5-way 5-shot | accuracy: 55.52 \pm 0.50\% -> 60.12 \pm 0.49\% | Helped |
| Add ALFA adaptive updates to MAML + L2F | MAML + L2F with 4-CONV · miniImageNet → FC100 5-way 5-shot | accuracy: 61.99 \pm 0.49\% -> 63.33 \pm 0.45\% | Helped |
Closest papers by meaning
- Unraveling Model-Agnostic Meta-Learning via The Adaptation Learning Rate (ICLR 2022 · similarity 0.90)
- Efficient Meta Learning via Minibatch Proximal Update (NeurIPS 2019 · similarity 0.90)
- MetaOptimize: A Framework for Optimizing Step Sizes and Other Meta-parameters (ICML 2025 · similarity 0.89)
- Memory-Reduced Meta-Learning with Guaranteed Convergence (AAAI 2025 · similarity 0.88)
- Provable Meta-Learning with Low-Rank Adaptations (NeurIPS 2025 · similarity 0.88)
- Delve into the Applicability of Advanced Optimizers for Multi-Task Learning (NeurIPS 2026 · similarity 0.87)
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.