SMS scnews item created by Garth Tarr at Thu 15 Nov 2018 1132
Type: Other
Distribution: World
Expiry: 13 Dec 2018
Auth: gartht@ (gtar4178) in SMS-WASM

Domestic PhD Scholarship: Deep Features for Outlier Detection

Call for PhD Student Applications - International Research Collaboration Project 

Project Title: Deep Features for Outlier Detection 

Area of Focus: Data Mining, Machine Learning, Statistical Learning 

Project Outline: Outlier (or anomaly) detection in data sets refers to the discovery of
patterns that do not conform to an established behaviour.  Relatively recently, the
problem of outlier detection has been considered in the context of deep learning
approaches.  The primary goal of this project is to reinvestigate the problem of
unsupervised outlier detection in light of recent advances in the unsupervised
generation of deep feature sets in machine learning.  

Prospective Students: Students with a strong CV and a passion for research and problem
solving using a combination of applied statistics/mathematics and computer science are
particularly encouraged to apply.  

International Research Team: 

James Bailey, University of Melbourne, Australia Ricardo J.  G.  B.  Campello,
University of Newcastle, Australia Michael E.  Houle, National Institute of Informatics,
Japan Arthur Zimek, University of Southern Denmark, Denmark 

Scholarship: At the moment, there is possibility of a scholarship available on a
competitive basis, for a domestic student in Australia.  A student granted with this
scholarship will be academically hosted mainly in the University of Newcastle, NSW,
Australia, but is expected to spend part of the PhD program as a visiting researcher in
the partner institutions.  

Kind regards, 


Ricardo J.  G.  B.  Campello | Professor – Data Science School of Mathematical &
Physical Sciences Room SR112, SR Building 

T: +61 2 4921 6762 E: W: 

The University of Newcastle University Drive, Callaghan NSW 2308 Australia

If you are registered you may mark the scnews item as read.
School members may try to .