ReinBo: Machine Learning pipeline search and configuration with Bayesian Optimization embedded Reinforcement Learning
This addresses the challenge of automated pipeline configuration for machine learning practitioners, though it appears incremental as it builds on existing optimization techniques.
The paper tackles the problem of optimizing hierarchical conditional hyper-parameter spaces in machine learning pipelines by proposing ReinBo, an algorithm combining Reinforcement Learning and Bayesian Optimization. Empirical results show it performs favorably compared to state-of-the-art methods like Auto-sklearn, TPOT, Tree Parzen Window, and Random Search.
Machine learning pipeline potentially consists of several stages of operations like data preprocessing, feature engineering and machine learning model training. Each operation has a set of hyper-parameters, which can become irrelevant for the pipeline when the operation is not selected. This gives rise to a hierarchical conditional hyper-parameter space. To optimize this mixed continuous and discrete conditional hierarchical hyper-parameter space, we propose an efficient pipeline search and configuration algorithm which combines the power of Reinforcement Learning and Bayesian Optimization. Empirical results show that our method performs favorably compared to state of the art methods like Auto-sklearn , TPOT, Tree Parzen Window, and Random Search.