TY - CHAP
T1 - A gradient-based forward greedy algorithm for space Gaussian process regression
AU - SUN, Ping
AU - YAO, Xin
PY - 2007
Y1 - 2007
N2 - In this chapter, we present a gradient-based forward greedy method for sparse approximation of Bayesian Gaussian Process Regression (GPR) model. Different from previous work, which is mostly based on various basis vector selection strategies, we propose to construct instead of select a new basis vector at each iterative step. This idea was motivated from the well-known gradient boosting approach. The resulting algorithm built on gradient-based optimisation packages incurs similar computational cost and memory requirements to other leading sparse GPR algorithms. Moreover, the proposed work is a general framework which can be extended to deal with other popular kernel machines, including Kernel Logistic Regression (KLR) and Support Vector Machines (SVMs). Numerical experiments on a wide range of datasets are presented to demonstrate the superiority of our algorithm in terms of generalisation performance.
AB - In this chapter, we present a gradient-based forward greedy method for sparse approximation of Bayesian Gaussian Process Regression (GPR) model. Different from previous work, which is mostly based on various basis vector selection strategies, we propose to construct instead of select a new basis vector at each iterative step. This idea was motivated from the well-known gradient boosting approach. The resulting algorithm built on gradient-based optimisation packages incurs similar computational cost and memory requirements to other leading sparse GPR algorithms. Moreover, the proposed work is a general framework which can be extended to deal with other popular kernel machines, including Kernel Logistic Regression (KLR) and Support Vector Machines (SVMs). Numerical experiments on a wide range of datasets are presented to demonstrate the superiority of our algorithm in terms of generalisation performance.
KW - Gaussian process regression
KW - sparse approximation
KW - sequential forward greedy algorithm
KW - basis vector selection
KW - basis vector construction
KW - gradient-based optimisation
KW - gradient boosting
UR - http://www.scopus.com/inward/record.url?scp=33750976960&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-36122-0_10
DO - 10.1007/978-3-540-36122-0_10
M3 - Book Chapter
SN - 9783540361213
T3 - Studies in Computational Intelligence
SP - 241
EP - 263
BT - Trends in Neural Computation
A2 - CHEN, Ke
A2 - WANG, Lipo
PB - Springer
ER -