TechRxiv
Online_Learning_with_RBF_Networks.pdf (1.46 MB)
Download file

Online Learning with Radial Basis Function Networks

Download (1.46 MB)
preprint
posted on 28.12.2021, 20:46 authored by Gabriel BorrageiroGabriel Borrageiro
We investigate the benefits of feature selection, nonlinear modelling and online learning when forecasting financial time series. We combine sequential updating with continual learning, specifically transfer learning. We perform feature representation transfer through clustering algorithms that determine the analytical structure of radial basis function networks we construct. These networks achieve lower mean-square prediction errors than kernel ridge regression models, which arbitrarily use all training vectors as prototypes. We also demonstrate quantitative procedures to determine the very structure of the networks. Finally, we conduct experiments on the log-returns of financial time series and show that these online transfer learning models outperform a random-walk baseline. In contrast, the offline learning models struggle to do so.

History

Email Address of Submitting Author

gabriel.borrageiro.20@ucl.ac.uk

ORCID of Submitting Author

https://orcid.org/0000-0002-0063-7103

Submitting Author's Institution

University College London

Submitting Author's Country

United Kingdom