Knowl Inf Syst (2014) 38:567–597
DOI 10.1007/s10115-012-0601-y
REGULAR PAPER
Transfer dimensionality reduction by Gaussian process
in parallel
Bin Tong · Junbin Gao · Thach Nguyen Huy ·
Hao Shao · Einoshin Suzuki
Received: 26 July 2011 / Revised: 8 August 2012 / Accepted: 19 December 2012 /
Published online: 8 January 2013
© Springer-Verlag London 2013
Abstract Dimensionality reduction has been considered as one of the most significant tools
for data analysis. In general, supervised information is helpful for dimensionality reduction.
However, in typical real applications, supervised information in multiple source tasks may
be available, while the data of the target task are unlabeled. An interesting problem of how to
guide the dimensionality reduction for the unlabeled target data by exploiting useful knowl-
edge, such as label information, from multiple source tasks arises in such a scenario. In this
paper, we propose a new method for dimensionality reduction in the transfer learning setting.
Unlike traditional paradigms where the useful knowledge from multiple source tasks is trans-
ferred through distance metric, we attempt to learn a more informative mapping function
between the original data and the reduced data by Gaussian process that behaves more appro-
priately than other parametric regression methods due to its less parametric characteristic.
This work is supported by the grant-in-aid for scientific research on fundamental research (B) 21300053
from the Japanese Ministry of Education, Culture, Sports, Science and Technology, and Charles Sturt
University Competitive Research Grant OPA 4818. This work was partially supported by the National
Science Foundation (Grant No. 61133016), and the National High Technology Joint Research Program of
China (863 Program, Grant No. 2011AA010706).
B. Tong (B ) · T. Nguyen Huy · H. Shao
Graduate School of Systems Life Sciences, Kyushu University, Fukuoka, Japan
e-mail: bintong@i.kyushu-u.ac.jp
T. Nguyen Huy
e-mail: thachnh@i.kyushu-u.ac.jp
H. Shao
e-mail: shaohao@i.kyushu-u.ac.jp
J. Gao
School of Computing and Mathematics, Charles Sturt University, Bathurst, Australia
e-mail: jbgao@csu.edu.au
E. Suzuki
Department of Informatics, ISEE, Kyushu University, Fukuoka, Japan
e-mail: suzuki@inf.kyushu-u.ac.jp
123