On The Relationship Between Online Gaussian Process Regression And Kernel Least Mean Squares Algorithms

TitleOn The Relationship Between Online Gaussian Process Regression And Kernel Least Mean Squares Algorithms
Publication TypeConference Paper
Year of Publication2016
AuthorsS. Van Vaerenbergh, J. Fernández-Bes, and V. Elvira
Conference Name2016 IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2016)
Month PublishedSeptember
PublisherIEEE
Conference LocationSalerno, Italy
Keywordsgaussian processes, kernel adaptive filtering, KLMS
AbstractWe study the relationship between online Gaussian process (GP) regression and kernel least mean squares (KLMS) algorithms. While the latter have no capacity of storing the entire posterior distribution during online learning, we discover that their operation corresponds to the assumption of a fixed posterior covariance that follows a simple parametric model. Interestingly, several well-known KLMS algorithms correspond to specific cases of this model. The probabilistic perspective allows us to understand how each of them handles uncertainty, which could explain some of their performance differences.
PDF version: