Hyperparameter Importance Across Datasets

TL;DR

This paper introduces a meta-learning framework to identify the most important hyperparameters across datasets and to infer good prior values, enhancing hyperparameter optimization for algorithms like SVMs, random forests, and Adaboost.

Contribution

It presents a novel methodology for automatic analysis of hyperparameter importance and priors using meta-data, advancing understanding beyond performance optimization.

Findings

Identified key hyperparameters for SVMs, random forests, and Adaboost.

Automatically inferred priors that improve hyperparameter optimization.

Validated the importance of selected hyperparameters through statistical significance.

Abstract

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.