Research map / Architectures and efficiency / Neural architecture search
PyGlove: Symbolic Programming for Automated Machine Learning
Daiyi Peng, Xuanyi Dong, Esteban Real, Mingxing Tan, Yifeng Lu, Gabriel Bender, Hanxiao Liu, Adam Kraft, Chen Liang, Quoc V. Le
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Search kernel size and expansion ratio with Random Search | MobileNetV2 · ImageNet validation | accuracy: 73.1 -> 73.7 | Improved |
| Use Bayesian search instead of Random Search on S1 | MobileNetV2 · ImageNet validation | accuracy: 73.7 -> 73.9 | Improved |
| Use TuNAS instead of Bayesian search on S1 | MobileNetV2 · ImageNet validation | accuracy: 73.9 -> 74.2 | Improved |
| Search output filters with TuNAS | MobileNetV2 · ImageNet validation | accuracy: 73.1 -> 73.3 | Improved |
| Combine kernel size, expansion ratio, and output-filter search | MobileNetV2 · ImageNet validation | accuracy: 73.1 -> 73.8 | Improved |
| Factorize search into node optimization followed by edge optimization | NAS-Bench-101 search with Regularized Evolution · NAS-Bench-101 | CIFAR-10 test accuracy | Hurt |
| Use a hybrid flow: optimize nodes first, then edges using the best node setting | NAS-Bench-101 search with Regularized Evolution · NAS-Bench-101 | CIFAR-10 test accuracy | Improved |
Closest papers by meaning
- Pipeline Combinators for Gradual AutoML (NeurIPS 2021 · similarity 0.88)
- AutoML-Zero: Evolving Machine Learning Algorithms From Scratch (ICML 2020 · similarity 0.86)
- An ADMM Based Framework for AutoML Pipeline Configuration (AAAI 2020 · similarity 0.86)
- AutoMMLab: Automatically Generating Deployable Models from Language Instructions for Computer Vision Tasks (AAAI 2025 · similarity 0.82)
- ADELA: Accelerating Evolutionary Design of Machine Learning Pipelines with the Accompanying Surrogate Model (AAAI 2025 · similarity 0.82)
- AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML (ICML 2025 · similarity 0.81)
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.
