Modelling nonstationary dynamics

We incorporate the use of validation data to cope with noisy records in a neural network-based method for modelling the dynamics of slowly changing nonstationary systems. As a byproduct, we obtain a precise criterion to find the optimal value of a required internal hyperparameter. Testing these idea...

Full description

Saved in:
Bibliographic Details
Published in:Physica A 2003-09, Vol.327 (1), p.190-194
Main Authors: Széliga, M.I., Verdes, P.F., Granitto, P.M., Ceccatto, H.A.
Format: Article
Language:English
Subjects:
Dynamical systems
Neural networks
Nonstationary time series
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We incorporate the use of validation data to cope with noisy records in a neural network-based method for modelling the dynamics of slowly changing nonstationary systems. As a byproduct, we obtain a precise criterion to find the optimal value of a required internal hyperparameter. Testing these ideas on a controlled problem shows that the resulting algorithm is able to outperform previous methods in the literature, allowing a more accurate modelling of nonstationary dynamics.
ISSN:0378-4371
1873-2119
DOI:10.1016/S0378-4371(03)00475-8