Research map / Architectures and efficiency / Neural architecture search
Meta Architecture Search
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 |
|---|---|---|---|
| Fine-tune the multi-task prior on CIFAR10 | BASE · CIFAR10 | Top-1 test error: 3.18 -> 2.83 | Helped |
| Fine-tune the multi-task prior on Imagenet32 | BASE · CIFAR10 | Top-1 test error: 3.18 -> 3.00 | Helped |
| Fine-tune the multi-task prior on SVHN | BASE · SVHN | Top-1 test error: 2.13 -> 2.01 | Helped |
| Fine-tune the multi-task prior on Imagenet32 | BASE · SVHN | Top-1 test error: 2.13 -> 2.07 | Helped |
| Fine-tune the multi-task prior on ImageNet | BASE · ImageNet | Top-1 error: 26.1 -> 25.7 | Helped |
| Fine-tune the multi-task prior on ImageNet | BASE · ImageNet | Top-5 error: 8.5 -> 8.1 | Helped |
| Use Gumbel-Softmax instead of softmax for architecture parameters (ablation) | BASE few-shot search · Mini-Imagenet 5-way 5-shot | 5-shot test accuracy: 65.4 ± 0.74% -> 66.2 ± 0.7% | Helped |
| Freeze the architecture parameters during adaptation (ablation) | BASE fast adaptation · CIFAR10 | accuracy | Hurt |
Closest papers by meaning
- Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets (ICLR 2021 · similarity 0.92)
- M-NAS: Meta Neural Architecture Search (AAAI 2020 · similarity 0.92)
- BayesNAS: A Bayesian Approach for Neural Architecture Search (ICML 2019 · similarity 0.92)
- Towards Fast Adaptation of Neural Architectures with Meta Learning (ICLR 2020 · similarity 0.92)
- Task-Adaptive Neural Network Search with Meta-Contrastive Learning (NeurIPS 2021 · similarity 0.92)
- Rapid Model Architecture Adaption for Meta-Learning (NeurIPS 2022 · similarity 0.91)
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.