Reinforcement-based Simultaneous Algorithm and its Hyperparameters Selection
This addresses the challenge of automating algorithm and hyperparameter selection for data analysis, particularly in classification, but is incremental as it builds on existing bandit-based methods.
The paper tackles the problem of simultaneously selecting a classification algorithm and its hyperparameters by reducing it to a multi-armed bandit problem, with experiments on 10 real datasets showing it is significantly better than Auto-WEKA in most cases and never worse.
Many algorithms for data analysis exist, especially for classification problems. To solve a data analysis problem, a proper algorithm should be chosen, and also its hyperparameters should be selected. In this paper, we present a new method for the simultaneous selection of an algorithm and its hyperparameters. In order to do so, we reduced this problem to the multi-armed bandit problem. We consider an algorithm as an arm and algorithm hyperparameters search during a fixed time as the corresponding arm play. We also suggest a problem-specific reward function. We performed the experiments on 10 real datasets and compare the suggested method with the existing one implemented in Auto-WEKA. The results show that our method is significantly better in most of the cases and never worse than the Auto-WEKA.