Hyperparameter Importance Across Datasets
This work provides a quantitative basis for focusing efforts in manual algorithm design and automated hyperparameter optimization, addressing a gap in automated analyses beyond performance optimization.
The paper tackled the problem of identifying generally important hyperparameters and good values for algorithms across datasets, using meta-learning on OpenML data for SVM, random forests, and Adaboost, resulting in statistically significant improvements in hyperparameter optimization.
With the advent of automated machine learning, automated hyperparameter optimization methods are by now routinely used in data mining. However, this progress is not yet matched by equal progress on automatic analyses that yield information beyond performance-optimizing hyperparameter settings. In this work, we aim to answer the following two questions: Given an algorithm, what are generally its most important hyperparameters, and what are typically good values for these? We present methodology and a framework to answer these questions based on meta-learning across many datasets. We apply this methodology using the experimental meta-data available on OpenML to determine the most important hyperparameters of support vector machines, random forests and Adaboost, and to infer priors for all their hyperparameters. The results, obtained fully automatically, provide a quantitative basis to focus efforts in both manual algorithm design and in automated hyperparameter optimization. The conducted experiments confirm that the hyperparameters selected by the proposed method are indeed the most important ones and that the obtained priors also lead to statistically significant improvements in hyperparameter optimization.