Deep learning scientist

Royal Meteorological Institute, Brussels, Belgium
Closing date: 17 June 2018

The field of deep learning has known a spectacular development in recent years [LeCun et al, 2015]. Contrary to the classical approach to data processing where a known model is appplied to input data in order to generate new output data, in the deep learning approach the model itself is derived by training it by large amounts of known pairs of input data and corresponding output data. The deep learning method is particularly well suited for application to problems where it is known that candidate input data contains information about a desired output, but where it is not straight forward to write down an explicit physical model desribing how the output data can be derived from the input data. The estimation of precipitation from multiple observation sources is such a problem. The scientist will work on the application of deep learning algorithms to the estimation of precipitation 1) from MSG SEVIRI satellite data, and 2) from dual polarisation radars.

Candidates should have the following qualifications:

  • an engineering master degree with scientific interest or scientific master degree with engineering interest
  • interest in making a PhD
  • interest/experience in deep learning techniques
  • interest/experience in atmospheric sciences in particular satelite and/or precipitation radar remote sensing
  • experience in programming
  • good knowledge of English
  • the knowledge of at least one of the national languages (Dutch and/or French) is an asset

The interested candidates for a long-term contract should sent their CV and motivation letter before 17 June 2018 to:

By continuing to use the site, you agree to our privacy policy and the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.