We invite applications for a fully-funded PhD research fellowship on the subject of deep neural networks and CRF normalization for graph-based syntactic dependency parsing.
Subject A detailed description of the subject can be found here: http://www-lipn.univ-paris13.fr/~leroux/parse-dnn.pdf
Eligibility and required skills
This proposal is intended for master students with one of the following backgrounds:
proficient in NLP and Computational Linguistics (CL) eager to discover
deep learning, or
proficient in machine learning with a strong motivation to learn NLP and
Fluency in English is a must.
The fellowship will be fully funded by Galilée graduate school ( http://www.univ-paris13.fr/ecole-doctorale-galilee) of University Paris 13 ( http://www.univ-paris13.fr), a member of Sorbonne Paris Cité ( http://www.sorbonne-paris-cite.fr).
The PhD will be funded for 3 years, with gross salary about 2000 euros/month (including teaching missions).
The PhD candidate will be working at LIPN (Laboratoire d'Informatique de Paris Nord - http://lipn.univ-paris13.fr) a CNRS UMR (7030). The PhD project is a collaboration between two teams:
RCLN (knowledge representation and natural language processing), a
member of Labex EFL excellence network (http://www.labex-efl.org); and
A3 (machine learning and applications).
The PhD candidate will be supervised by Yann Chevaleyre (Full Professor, ML, http://lipn.univ-paris13.fr/~chevaleyre), Joseph Le Roux (Assistant Professor, NLP, http://lipn.univ-paris13.fr/~leroux) and Nadi Tomeh (Assistant Professor, NLP, http://lipn.univ-paris13.fr/~tomeh).
How to Apply
Please send your application to Joseph Le Roux (leroux at univ-paris13.fr). Applications must contain:
Degrees and grades, especially for the Masters (M1 and M2 in France);
The deadline for applications is 25/5/2016.
This deadline is firm and no applications can be received beyond it.
Starting date September 2016.
-- Nadi Tomeh LIPN, Université Paris 13 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 14316 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20160519/a9e90a97/attachment.txt>