Initialization of multilayer forecasting artifical neural networks

TL;DR

This paper introduces a novel initialization method for multilayer neural networks predicting time series, inspired by linear prediction filters, and demonstrates its effectiveness on chaotic system forecasting.

Contribution

The paper presents a new initialization technique for multilayer neural networks based on linear prediction filters and matrix decomposition, improving forecasting accuracy.

Findings

Effective neural network initialization for time series prediction.

Improved accuracy in forecasting chaotic systems.

Validation on Lorentz system demonstrates method's utility.

Abstract

In this paper, a new method was developed for initialising artificial neural networks predicting dynamics of time series. Initial weighting coefficients were determined for neurons analogously to the case of a linear prediction filter. Moreover, to improve the accuracy of the initialization method for a multilayer neural network, some variants of decomposition of the transformation matrix corresponding to the linear prediction filter were suggested. The efficiency of the proposed neural network prediction method by forecasting solutions of the Lorentz chaotic system is shown in this paper.