An ADMM Based Framework for AutoML Pipeline Configuration
This addresses the challenging problem of automated pipeline configuration for practitioners, though it appears incremental as it builds on existing AutoML techniques with a novel optimization approach.
The authors tackled the AutoML problem of configuring machine learning pipelines by jointly selecting algorithms and hyperparameters, proposing an ADMM-based framework that decomposes the mixed-variable optimization into simpler sub-problems and handles black-box constraints. They empirically showed their framework provides significant gains compared to Auto-sklearn and TPOT on binary classification datasets.
We study the AutoML problem of automatically configuring machine learning pipelines by jointly selecting algorithms and their appropriate hyper-parameters for all steps in supervised learning pipelines. This black-box (gradient-free) optimization with mixed integer & continuous variables is a challenging problem. We propose a novel AutoML scheme by leveraging the alternating direction method of multipliers (ADMM). The proposed framework is able to (i) decompose the optimization problem into easier sub-problems that have a reduced number of variables and circumvent the challenge of mixed variable categories, and (ii) incorporate black-box constraints along-side the black-box optimization objective. We empirically evaluate the flexibility (in utilizing existing AutoML techniques), effectiveness (against open source AutoML toolkits),and unique capability (of executing AutoML with practically motivated black-box constraints) of our proposed scheme on a collection of binary classification data sets from UCI ML& OpenML repositories. We observe that on an average our framework provides significant gains in comparison to other AutoML frameworks (Auto-sklearn & TPOT), highlighting the practical advantages of this framework.