High Per Parameter: A Large-Scale Study of Hyperparameter Tuning for Machine Learning Algorithms

TL;DR

This large-scale study evaluates the actual benefits of hyperparameter tuning across 26 ML algorithms and 250 datasets, revealing that tuning often yields limited improvements but can be crucial for certain datasets and algorithms.

Contribution

It provides the first comprehensive large-scale analysis of hyperparameter tuning effectiveness across diverse algorithms and datasets, introducing a ranking method based on a new hp_score metric.

Findings

Many algorithms show limited gains from tuning on average.

Some datasets benefit significantly from hyperparameter tuning.

A ranking of algorithms by potential tuning benefit is proposed.

Abstract

Hyperparameters in machine learning (ML) have received a fair amount of attention, and hyperparameter tuning has come to be regarded as an important step in the ML pipeline. But just how useful is said tuning? While smaller-scale experiments have been previously conducted, herein we carry out a large-scale investigation, specifically, one involving 26 ML algorithms, 250 datasets (regression and both binary and multinomial classification), 6 score metrics, and 28,857,600 algorithm runs. Analyzing the results we conclude that for many ML algorithms we should not expect considerable gains from hyperparameter tuning on average, however, there may be some datasets for which default hyperparameters perform poorly, this latter being truer for some algorithms than others. By defining a single hp_score value, which combines an algorithm's accumulated statistics, we are able to rank the 26 ML algorithms from those expected to gain the most from hyperparameter tuning to those expected to gain the least. We believe such a study may serve ML practitioners at large.