School of Informatics, University of Edinburgh
Research Position in Statistical Machine Translation
This project deals with applying large-scale discriminative techniques
(such as tree-based Conditional Random Fields or Perceptrons) for
Statistical Machine Translation. It also is concerned with
understanding how linguistic resources can (or cannot) be usefully
incorporated into such translation models. Significant challenges
include designing training regimes
which scale with very large quantities of data, how to regularise such models
and how to combine insights from syntax-based Machine Translation with
Phrase-based Machine Translation.
The ideal candidate will hold a PhD (or equivalent) in Computer
Science or some other numerate subject. Strong programming ability
(ideally in C++) and a good understanding of discriminative machine
learning will be highly advantageous. The candidate will join the
Edinburgh Machine Translation Group and it is anticipated entering
into the annual NIST Machine Translation evaluation exercises.
The post is for 3 years and further details can be found online:
http://www.jobs.ed.ac.uk/vacancies/index.cfm?fuseaction=vacancies.detail&vacancy_ref=3006168
Closing date: 08-Aug-2006
The Edinburgh Machine Translation Group:
http://forbes.inf.ed.ac.uk/ued/
ICCS:
http://www.iccs.informatics.ed.ac.uk/
The School of Informatics:
http://www.informatics.ed.ac.uk/
Informal enquiries may be made to:
Miles Osborne (miles@inf.ed.ac.uk)
http://www.cogsci.ed.ac.uk/~osborne/
This archive was generated by hypermail 2b29 : Tue Jul 18 2006 - 23:56:09 MET DST