Hyperparameter Importance for Machine Learning Algorithms

Abstract

Hyperparameter plays an essential role in the fitting of supervised machine learning algorithms. However, it is computationally expensive to tune all the tunable hyperparameters simultaneously especially for large data sets. In this paper, we give a definition of hyperparameter importance that can be estimated by subsampling procedures. According to the importance, hyperparameters can then be tuned on the entire data set more efficiently. We show theoretically that the proposed importance on subsets of data is consistent with the one on the population data under weak conditions. Numerical experiments show that the proposed importance is consistent and can save a lot of computational resources.

A Comparative Analysis of XGBoost and Neural Network Models for Predicting Some Tomato Fruit Quality Traits from Environmental and Meteorological Data

“…The lack of variation in the direction of SHAP values for the ANN model contrasted sharply with the XGBoost model and suggested that the former may not be capturing the true underlying data patterns. This discrepancy emphasizes the need for a thorough cross-validation and hyperparameter tuning process, which has been identified as a crucial step in model development [ 113 , 114 ]. Furthermore, the negative R² score suggests that the ANN model’s predictive power was worse than a naïve model, which would simply predict the average a/b ratio for all observations [ 115 ].…”