FLAML: A Fast and Lightweight AutoML Library
arXiv:1911.04706
Abstract
We study the problem of using low computational cost to automate the choices of learners and hyperparameters for an ad-hoc training dataset and error metric, by conducting trials of different configurations on the given training data. We investigate the joint impact of multiple factors on both trial cost and model error, and propose several design guidelines. Following them, we build a fast and lightweight library FLAML which optimizes for low computational resource in finding accurate models. FLAML integrates several simple but effective search strategies into an adaptive system. It significantly outperforms top-ranked AutoML libraries on a large open source AutoML benchmark under equal, or sometimes orders of magnitude smaller budget constraints.
14 pages, published in Fourth Conference on Machine Learning and Systems (MLSys 2021)
References in corpus (6)
- Practical Bayesian Optimization of Machine Learning Algorithms
- Neural Architecture Search: A Survey
- OpenML: networked science in machine learning
- Learned Cardinalities: Estimating Correlated Joins with Deep Learning
- AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data
- Cloudy with high chance of DBMS: A 10-year prediction for Enterprise-Grade ML
Cited by in corpus (4)
- Photometric redshift-aided classification using ensemble learning
- Min-Max Optimization without Gradients: Convergence and Applications to Adversarial ML
- Automated classification of Chandra X-ray point sources using machine learning methods
- Meta-Learning for Unsupervised Outlier Detection with Optimal Transport