On-Line Learning Methods for Gaussian Processes

Shigeyuki Oba, Masa-aki Sato, Shin IshiiIEICE 2003

Abstract

This article proposes two modifications of Gaussian processes, which aim to deal with dynamic environments. One is a weight decay method that gradually forgets the old data, and the other is a time stamp method that regards the time course of data as a Gaussian process. We show experimental results when these modifications are applied to regression problems in dynamic environments.
BibTeX
@inproceedings{oba2003line,
  author = {Shigeyuki Oba and Masa-aki Sato and Shin Ishii},
  title = {{On-Line Learning Methods for Gaussian Processes}},
  booktitle = {IEICE},
  year = {2003},
}

Loading related content...

Figure