PhD Position in Computer Science

University of Bern Institute of Computer Science

Switzerland

PhD Position in Computer Science
Project on Graph Based Methods for Pattern Recognition and Machine Learning

We are seeking a motivated PhD candidate for our SNF-funded research project "Novel State-of-the-Art Graph Matching Algorithms".

In this project we plan to make significant advances in the field of structural pattern recognition and establishing novel paradigms that go beyond the current understanding. In particular, the overall objective is the development and research of novel, robust graph edit distance methods that outperform the current state-of-the-art in graph matching on existing and novel data sets stemming from different real world scenarios. Hence, the proposed project involves both research on fundamental algorithms and solving concrete problems in applications.

Research

The PhD candidate will research hierarchical graph representations in conjunction with linear time graph embedding. This procedure is motivated by the fact that hierarchical representations (including fast and expressive graph embeddings) can be exploited in filter-and-verify strategies in order to substantially speed up and improve the matching processes.
Requirements
The PhD candidate should have a masters degree in computer science.

The ideal candidate:
  • Has a strong interest and background in pattern recognition and graph based representation
  • Has excellent programming skills (in particular Python and/or Java)
  • Has excellent writing and presentation skills
Interested?

For further information please contact the PhD supervisor: kaspar.riesen@inf.unibe.ch
The applicant should submit a CV (including contacts of two referees) and a one-page motivation letter, highlighting her/his experiences in the above-mentioned points and indicating his/her motivation to join the project.

Please send your application electronically to: kaspar.riesen@inf.unibe.ch


In your application, please refer to Polytechnicpositions.com

FACEBOOK
TWITTER
LINKEDIN

baner1

baner10

baner12

baner14

baner2

baner3

baner4

baner5

baner6

baner7

baner8

baner9