Pupil Learning Mechanism

TL;DR

This paper introduces the Pupil Learning Mechanism (PLM), a novel approach for modifying neural network structures and weights to address vanishing gradients and overfitting, validated through experiments on copper price forecasting.

Contribution

The paper proposes the PLM framework with modules for sequential, adaptive, perfect, and less-overfitted learning, advancing neural network training methods.

Findings

PLM modules are effective in improving learning performance.

PLM outperforms linear regression and traditional backpropagation models.

Empirical validation confirms the superiority of PLM in forecasting tasks.

Abstract

Studies on artificial neural networks rarely address both vanishing gradients and overfitting issues. In this study, we follow the pupil learning procedure, which has the features of interpreting, picking, understanding, cramming, and organizing, to derive the pupil learning mechanism (PLM) by which to modify the network structure and weights of 2-layer neural networks (2LNNs). The PLM consists of modules for sequential learning, adaptive learning, perfect learning, and less-overfitted learning. Based upon a copper price forecasting dataset, we conduct an experiment to validate the PLM module design modules, and an experiment to evaluate the performance of PLM. The empirical results indeed approve the PLM module design and show the superiority of the proposed PLM model over the linear regression model and the conventional backpropagation-based 2LNN model.