¡¡
RESEARCH
ACTIVITIES
Research Interests:
Machine Learning (e.g., Deep Learning and Transfer Learning),
Evolutionary Optimization (e.g. Differential Evolution and Particle
Swarm Optimization), Image Processing (e.g., Remote Sensing Image
Processing and Automatic Optical Inspection), GPU Computing, Services
Computing (e.g., Service Composition), Mobile and Pervasive Computing,
etc.
My current research focuses on
Collaborative
Learning and Optimization, i.e., how to synergize machine learning
and intelligent optimization techniques to resolve challenging problems
in which learning and optimization are involved as indispensable and
interwoven tasks.
Research Grants:
-
Research on Theory and Algorithms in Multi-Objective Evolutionary Learning for Submodular Contrained Optimization (2017-2019), National Natural Science Foundation of
China (NSFC), £¤220,000 (Co-PI)
-
Long-term Cloud
Service Composition (2016-2018), Australia Research Council (ARC)
Discovery Project, AU$339,000 (CI)
-
Research grant for
the recipient of the 2014 Chinese National 1000 Young Talents
Plan,£¤3,000,000 (CI)
-
Novel Adaptive
Ensemble Evolutionary Optimization Techniques (2011-2013), National
Natural Science Foundation of China (NSFC),£¤200,000 (CI)
-
Investigation of
Ensemble Evolutionary Algorithms and Its Applications (2011-2013),
Specialized Research Fund for the Doctoral Program of Higher
Education (SRFDP), £¤36,000 (CI)
Past Research Projects:
-
Post-reflow defect detection in SMT
(Surface-Mounted Technology)
Funding sources:
I-VP: Intuitive Vision Programming (MINALOGIC)
(French
Government Funding), 2009-2012
-
Remote sensing imagery intepretation
Funding sources:
Classification of Operational SAR Sea
Ice Imagery (NSERC-Discovery Funding), 2007-2012; Variability and Change in the Canadian Cryosphere¨CCanadian
Contribution to the State and Fate of the Polar Cryosphere (Environment
Canada Funding), 2007-2011; Classification of Operational SAR Sea Ice
Imagery (GEOIDE SII Funding), 2006-2008
MAGIC (MAp-Guided
Ice Classification)
is a software package that interprets remote sensing imagery using
advanced computer vision techniques, which feature an innovative
segmentation algorithm using probabilistic graphical models to
reducesegmentation errors due to signal variations in the range
direction and speckle noise.
CURRENT PHD
STUDENTS
|