[visionlist] Technical University of Munich: Two full-time PhD positions on Causality Modeling and Incremental Learning

[Marco Körner]

The Chair of Remote Sensing Technology at Technical University of Munich
(TUM), Germany, invites applications to fill two full-time (100%) PhD
positions (TV-L E13) in the recently granted Horizon 2020 Project „Global
Earth Monitor“ (GEM), with an immediate start.
Within this interdisciplinary project with partners from academia and
industry, TUM will push forward the development of machine learning models
to extract rich information from massive volumes of optical Earth
observation imagery, especially taking their multi-spatial, multi-spectral,
and multi-temporal nature into account.

For the full job openings, please refer to
https://portal.mytum.de/jobs/wissenschaftler/NewsArticle_20201106_133543
and
https://portal.mytum.de/jobs/wissenschaftler/NewsArticle_20201106_154005


We are searching for applicants for the following research topics:

#1 Causality Modeling and Change Detection
to derive the driving causal relationships between individual observations
and processes, as well as to recognize and evaluate change processes on the
Earth’s surface

#2 Incremental Learning
to update learned models after obtaining new observations without
re-training


Duties:
- conducting basic and methodological research within the scope of GEM
- analysis of optical Earth observation data
- implementation of the developed algorithms and their publication
- experimental evaluation of machine learning methods for Earth observation
data analysis
- supporting our teaching activities

Requirements:
- excellently-qualified graduates with a diploma or master's degree in
computer science, electrical engineering, data science, geoinformatics,
statistics, mathematics, physics, or comparable fields
- in-depth methodological knowledge in the field of machine learning,
especially in the areas of
  #1: change detection & interpretation, causality modeling, and
uncertainty assessment
  #2: incremental and curriculum learning; multi-task, multi-object &
meta-learning; and time series and multi-scale analysis
- interest in working with multi-modal, multi-scalar, and multi-temporal
Earth observation data
- practical experience with software development in Python and the common
deep learning frameworks and libraries
- analytical thinking, independent and structured work, as well as the
willingness to cooperate with other team members
- good English language skills

Our offer:
- a full-time position (100%) for initially up to 28 months with an option
for extension and remuneration according to TV-L up to E13
- the opportunity and support to pursue a structured doctoral degree at one
of the most renowned universities in Europe with international exchange and
direct practical relevance through cooperation with partners
- a dynamic working environment in a young, curiosity-driven, and
research-interested team

Application
We look forward to receiving your application documents (cover letter,
curriculum vitae, study certificates, ...) by 30.11.2020, preferably via
e-mail to apply at lmf.lrg.tum.de. Application documents sent by post will not
be returned after the procedure is completed. If you have any questions
regarding the position, please contact Ms. Maja Schneider (
maja.schneider at tum.de) or Dr. Marco Körner (marco.koerner at tum.de).


-- 
Dr. Marco Körner

Deputy Head
Chair of Remote Sensing Technology

Group Leader
Computer Vision Research Group

Technical University of Munich
Department of Aerospace and Geodesy
Arcisstr. 21, 80333 München, Germany

Mail: marco.koerner at tum.de
Tel.: +49-89-289 22674
WWW : www.lmf.lrg.tum.de
      www.lmf.lrg.tum.de/vision
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20201111/2c71544c/attachment.html>