Online Lazy Gaussian Process Committee and its Application in Real-Time Trajectory Prediction
A significant problem of Gaussian process (GP) is its unfavorable scaling with a large amount of data. To overcome this issue, we present a novel GP approximation scheme for online regression. Our model is based on a combination of multiple GPs with random hyperparameters. The model is trained by incrementally allocating new examples to a selected subset of GPs. The selection is carried out efficiently by optimizing a submodular function. Experiments on real-world data sets showed that our method outperforms existing online GP regression methods in both accuracy and efficiency. The applicability of the proposed method is demonstrated by the mouse-trajectory prediction in an Internet banking scenario.
Online Lazy Gaussian Process Committee and its Application in Real-Time Trajectory Prediction
Information Sciences
Authors: | Han Xiao and Claudia Eckert |
Year/month: | 2013/12 |
Note: | (IF: 3.676) Accepted, to appear |
Fulltext: | draft20131014.pdf |
Abstract |
|
A significant problem of Gaussian process (GP) is its unfavorable scaling with a large amount of data. To overcome this issue, we present a novel GP approximation scheme for online regression. Our model is based on a combination of multiple GPs with random hyperparameters. The model is trained by incrementally allocating new examples to a selected subset of GPs. The selection is carried out efficiently by optimizing a submodular function. Experiments on real-world data sets showed that our method outperforms existing online GP regression methods in both accuracy and efficiency. The applicability of the proposed method is demonstrated by the mouse-trajectory prediction in an Internet banking scenario. |
Bibtex:
@article { hanxiao2013d,author = { Han Xiao and Claudia Eckert},
title = { Online Lazy Gaussian Process Committee and its Application in Real-Time Trajectory Prediction },
journal = { Information Sciences },
year = { 2013 },
month = { December },
note = { (IF: 3.676) Accepted, to appear },
url = {https://www.sec.in.tum.de/i20/publications/online-lazy-gaussian-process-committee-and-its-application-in-real-time-trajectory-prediction/@@download/file/draft20131014.pdf}
}