Distance Searching-based Hyperparameter Optimization for Restricted Coulomb Energy-based Neural Network
Kyou Jung Son, Seokhun Jeon, Jae-Hack Lee, Byung‐Soo Kim · 2023
In this paper, we propose an optimal hyperparameter searching algorithm for the restricted coulomb energy (RCE) network-based classification algorithm. The proposed algorithm analyzes the input data to compute the distances between their features belonging to different classes. Based on these computed distances, the algorithm obtains the minimum and maximum distance values, which are then set as hyperparameters. To evaluate the performance of the proposed algorithm, we conducted experiments using benchmark datasets and compared its performance against a full search approach for hyperparameter settings. The results demonstrate that the proposed algorithm outperforms the full search method.