This PhD position targets the research into data efficient and transparent deep learning methods matching and recognition of 3D objects based on their 2d representations. The topics of interest include:

use of simulated data for investigating the feasibility of the approach for 3D reconstruction from 2D images in the absence of labelled real data;

semi-supervised approaches for learning latent representations for data efficient mapping of 2D to 3D geometry;

interpretable models explaining their reasoning for verifiability in copyright protection of 3D models.

The results of this PhD project will be validated in Naspers media and classified advertising platforms. The work will be carried out at the System Architecture and Networking group (the candidate will be hosted at Naspers office in Eindhoven), and will be funded by the Dutch program on 'Efficient Deep Learning' (www.stw.nl/nl/content/p16-25-efficient-deep-learning).

We are looking for candidates that meet the following requirements:

a master's degree in statistics or computer science (preferably with a specialization in artificial intelligence and/or machine learning),

specialization in deep learning, machine learning, data mining or related areas,

affinity with computational statistics or machine learning and excellent programming skills,

have machine learning/data mining software development skills in at least in one language, e.g. R, Java, Python (familiarity with deep learning libraries and Python is a plus),

ability to combine different research techniques and tools,

an excellent command of English and good academic writing and presentation skills

The PhD candidate is expected to:

perform scientific research in the domain described,

collaborate with other researchers in our faculty and the project,

transfer knowledge to internal specialists at Naspers,

publish results in leading scientific journals and present results at international conferences in the field,

assist in guiding MSc graduation projects.

full-time employment as a PhD-candidate for a period of 4 years;

annually 8% holiday allowance and 8.3% end of year allowance;

support with your personal development and career planning including courses, summer schools, conference visits etc.;

a broad package of fringe benefits (including an excellent technical infrastructure, child care, moving expenses, savings schemes, coverage of costs of publishing the dissertation and excellent sports facilities)