Efficient Hyperparameter Importance Assessment for CNNs

TL;DR

This paper introduces an efficient method for assessing hyperparameter importance in CNNs, helping practitioners focus on impactful parameters to improve model performance and reduce computational costs.

Contribution

It proposes the N-RReliefF algorithm for hyperparameter importance assessment and provides extensive empirical analysis across multiple datasets to identify key hyperparameters in CNNs.

Findings

Top five important hyperparameters identified: convolutional layers, learning rate, dropout, optimizer, epochs.

Demonstrated that hyperparameter importance varies across datasets and models.

Method reduces search space, saving time and computational resources.

Abstract

Hyperparameter selection is an essential aspect of the machine learning pipeline, profoundly impacting models' robustness, stability, and generalization capabilities. Given the complex hyperparameter spaces associated with Neural Networks and the constraints of computational resources and time, optimizing all hyperparameters becomes impractical. In this context, leveraging hyperparameter importance assessment (HIA) can provide valuable guidance by narrowing down the search space. This enables machine learning practitioners to focus their optimization efforts on the hyperparameters with the most significant impact on model performance while conserving time and resources. This paper aims to quantify the importance weights of some hyperparameters in Convolutional Neural Networks (CNNs) with an algorithm called N-RReliefF, laying the groundwork for applying HIA methodologies in the Deep Learning field. We conduct an extensive study by training over ten thousand CNN models across ten popular image classification datasets, thereby acquiring a comprehensive dataset containing hyperparameter configuration instances and their corresponding performance metrics. It is demonstrated that among the investigated hyperparameters, the top five important hyperparameters of the CNN model are the number of convolutional layers, learning rate, dropout rate, optimizer and epoch.