Interval-valued Data Prediction via Regularized Artificial Neural Network

TL;DR

This paper introduces a regularized artificial neural network (RANN) for interval-valued data prediction, incorporating a non-crossing regularizer to ensure mathematical coherence and demonstrating superior accuracy over traditional models.

Contribution

The paper proposes a novel RANN model with a soft non-crossing regularizer for interval data prediction, enhancing prediction accuracy and coherence.

Findings

RANN outperforms traditional models in prediction accuracy.

The non-crossing regularizer effectively maintains interval coherence.

Extensive experiments validate RANN's effectiveness on real and simulated data.

Abstract

A regularized artificial neural network (RANN) is proposed for interval-valued data prediction. The ANN model is selected due to its powerful capability in fitting linear and nonlinear functions. To meet mathematical coherence requirement for an interval (i.e., the predicted lower bounds should not cross over their upper bounds), a soft non-crossing regularizer is introduced to the interval-valued ANN model. We conduct extensive experiments based on both simulation datasets and real-life datasets, and compare the proposed RANN method with multiple traditional models, including the linear constrained center and range method (CCRM), the least absolute shrinkage and selection operator-based interval-valued regression method (Lasso-IR), the nonlinear interval kernel regression (IKR), the interval multi-layer perceptron (iMLP) and the multi-output support vector regression (MSVR). Experimental results show that the proposed RANN model is an effective tool for interval-valued prediction tasks with high prediction accuracy.