AISB Convention 2015

Call for Symposium Proposals: The AISB Convention is an annual conference covering the range of AI and Cognitive Science, organised by the Society for the Study of Artificial Intelligence and Simulation of Behaviour. The 2015 Con...


Read More...

Yasemin Erden on BBC

AISB Committee member, and Philosophy Programme Director and Lecturer, Dr Yasemin J. Erden interviewed for the BBC on 29 October 2013. Speaking on the Today programme for BBC Radio 4, as well as the Business Report for BBC world N...


Read More...

Mark Bishop on BBC ...

Mark Bishop, Chair of the Study of Artificial Intelligence and the Simulation of Behaviour, appeared on Newsnight to discuss the ethics of ‘killer robots’. He was approached to give his view on a report raising questions on the et...


Read More...

AISB YouTube Channel

The AISB has launched a YouTube channel: http://www.youtube.com/user/AISBTube (http://www.youtube.com/user/AISBTube). The channel currently holds a number of videos from the AISB 2010 Convention. Videos include the AISB round t...


Read More...

Lighthill Debates

The Lighthill debates from 1973 are now available on YouTube. You need to a flashplayer enabled browser to view this YouTube video  


Read More...
01234

Notice

AISB opportunities Bulletin Item

Postdoctoral Research Position, DIKU, University of Copenhagen


Contact: loog@diku.dk

Announcement Postdoctoral Research Position, DIKU, University of Copenhagen

The Image Group at the Department of Computer Science [DIKU], University 
of Copenhagen, Copenhagen, Denmark, has an opening for a postdoctoral 
researcher on the project "Learning in Multiscale Image Analysis". The 
postdoc is expected to develop [statistical] theory, and related 
[computational] algorithms, that advance the integration of multiscale 
image analysis methods and machine learning in a principled way.  If of 
interest, specific applications could be considered in more detail.

The initial position is for one year but may be extended by 1.5 years 
depending on performance and based on application.


Qualifications

The candidate should have either a strong background in [fundamental] 
multiscale image analysis or computer vision and have affinity with 
machine learning or pattern recognition or vice versa. Additionally, 
[s]he should feel comfortable with mathematics and statistics.  The 
applicant should have good demonstrated skills in written and spoken 
English.  A Ph.D. in a relevant and/or related discipline is a necessity.


Short Project Description

Various learning techniques from statistics, machine learning, and 
pattern recognition have been employed successfully to tackle 
challenging problems in image segmentation, image de-noising, filtering 
and related tasks.  These supervised techniques have been applied to 
such image problems in a rather straightforward and ad hoc manner, 
solving the task at hand, but providing no additional insight beyond 
this.  That is, image analysis tasks are solved in isolation and solving 
one problem will not necessarily provide any valuable clues on how to 
approach yet another problem.  We intend to investigate image analysis 
in combination with supervised methods from a more general perspective.

A thorough understanding of the interrelation between supervision and 
image analysis is a necessity for exploiting their combination to the 
full extent.  A principled connection between such technique, however, 
has not yet been established and the project intends to provide [first] 
steps towards such connection and its deeper understanding.  The focus 
will be on multiscale image representations, particularly, those 
identified with scale space theory.  Using this, based on methods from 
pattern recognition and machine learning and techniques from, e.g., 
multivariate statistics, differential geometry, and topology, the aim is 
to study and uncover general principles, laws, and concepts that govern 
supervised multiscale methods and employ such knowledge in specific 
algorithms in order to demonstrate proof of concept.


Contact Persons

Marco Loog, Ph.D.
Department of Computer Science
University of Copenhagen
Copenhagen, Denmark
loog@diku.dk

Prof. Mads Nielsen, Ph.D.
Department of Computer Science
University of Copenhagen
Copenhagen, Denmark
madsn@diku.dk