Research map / Learning theory and algorithms / Combinatorial and search optimization
Transferable AutoML by Model Sharing Over Grouped Datasets
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 |
|---|---|---|---|
| Share models across datasets using benchmark-performance grouping with Markov analysis and hypothesis testing | Hyperband · 7 image-classification datasets, search from scratch | search time (days): 10.40 -> 3.17 | Helped |
| Share models across datasets using benchmark-performance grouping with Markov analysis and hypothesis testing | Hyperband · 7 image-classification datasets, search from scratch | TRE: 0 -> 0.062 | Hurt |
| Share models across datasets using benchmark-performance grouping with Markov analysis and hypothesis testing | MetaQNN · 7 image-classification datasets, search from scratch | search time (days): 16.29 -> 4.85 | Helped |
| Share models across datasets using benchmark-performance grouping with Markov analysis and hypothesis testing | MetaQNN · 7 image-classification datasets, search from scratch | TRE: 0 -> 0.039 | Hurt |
| Share models across datasets using benchmark-performance grouping with Markov analysis and hypothesis testing | ENAS · 7 image-classification datasets, search from scratch | search time (days): 12.22 -> 4.22 | Helped |
| Share models across datasets using benchmark-performance grouping with Markov analysis and hypothesis testing | ENAS · 7 image-classification datasets, search from scratch | TRE: 0 -> 0.019 | Hurt |
| Share searched models across grouped datasets | MetaQNN · 4 image-classification datasets, search from pre-defined model | search time (days): 8 -> 3.0 | Helped |
| Share searched models across grouped datasets | MetaQNN · 4 image-classification datasets, search from pre-defined model | TRE: 0 -> 0.074 | Hurt |
| Transfer a basic cell from a grouped source dataset | Tr-AutoML (MH) · FASHION-MNIST | search time (days): 1.67 -> 0.013 | Helped |
| Transfer a basic cell from a grouped source dataset | Tr-AutoML (MH) · FASHION-MNIST | accuracy: 0.942 -> 0.939 | Hurt |
| Transfer a basic cell from a grouped source dataset | Tr-AutoML (MH) · STL10 | search time (days): 1.08 -> 0.017 | Helped |
| Transfer a basic cell from a grouped source dataset | Tr-AutoML (MH) · STL10 | accuracy: 0.734 -> 0.725 | Hurt |
Closest papers by meaning
- Transfer Learning with Neural AutoML (NeurIPS 2018 · similarity 0.91)
- AutoTransfer: AutoML with Knowledge Transfer - An Application to Graph Neural Networks (ICLR 2023 · similarity 0.87)
- AutoMMLab: Automatically Generating Deployable Models from Language Instructions for Computer Vision Tasks (AAAI 2025 · similarity 0.86)
- AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML (ICML 2025 · similarity 0.84)
- MLZero: A Multi-Agent System for End-to-end Machine Learning Automation (NeurIPS 2025 · similarity 0.82)
- AutoAL: Automated Active Learning with Differentiable Query Strategy Search (ICML 2025 · similarity 0.82)
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.