PhD position (m/f/d; E13 TV-L)

Mathematisch-Naturwissenschaftliche Fakultät, Fachbereich Informatik

Bewerbungsfrist : 30.09.2021

The Bosch Industry-on-Campus Lab, a research collaboration between the University of Tübingen and Bosch Center for AI (BCAI), invites applications for an open

PhD position (m/f/d; E13 TV-L)

in Generative Models to Improve Out-of-distribution Generalization. The assumption that the training and test data of machine learning systems are independently sampled from the same distribution does not necessarily hold in practice. Being able to trust model predictions at test time, the requirement on model’s generalization shall not only concern in-distribution behind the training data but also out-of-distributions beyond it. The aim of this project is to improve the out-of-distribution generalization of machine learning systems. To this end, you will specifically investigate deep generative models, exploiting their use for data manipulation, data generation recovery, representation learning, and density estimation. 

You will be jointly supervised by Dr. Dan Zhang from the Bosch side and by Prof. Andreas Geiger from the University of Tübingen side.

The position is available immediately (but start date is negotiable), the contract is initially for three years, and remunerated according to the German salary scale 13 TVL.

What are you going to do?

As part of the Bosch Industry-on-Campus Lab at the University of Tübingen, you are going to carry out AI research and develop novel machine learning methods for safety critical applications in autonomous driving, healthcare, and manufacturing systems. There will also be regular visits and interactions with researchers at Bosch Center for AI, who have an office on campus. At the University of Tübingen you will be supervised by Dr. Dan Zhang and Prof. Dr. Andreas Geiger.

Your tasks will be to:

  • Develop new generative methods to improve the out-of-distribution generalization of machine learning models;
  • Collaborate with other researchers within the lab and BCAI Research;
  • Complete and defend a PhD thesis within the official appointment duration of three years;
  • Regularly present internally on your progress and help Bosch write patent applications to protect inventions from the lab when requested.
  • Regularly present intermediate research results at international conferences and workshops, and publish them in proceedings and journals;
  • Potentially assist in relevant teaching activities.

What do we require?

  • Master’s degree in Computer Science, Artificial Intelligence, Mathematics, or related field;
  • Strong background in machine learning and/or computer vision;
  • Excellent programming skills, preferably in Python;
  • Prior experience of working with deep learning libraries, such as PyTorch or TensorFlow;
  • Solid mathematics foundations, especially in probability theory, statistics, calculus and linear algebra;
  • High motivation and creativity;
  • Strong communication, presentation and writing skills and excellent command of English.

Prior publications in relevant machine learning and vision venues as well as experience working with deep generative models will be advantageous for your application.


The University of Tübingen is an equal-opportunity employer. We prioritize diversity and are committed to creating an inclusive environment for everyone. We seek to increase diversity and the number of women in areas where they are under-represented and therefore explicitly encourage women to apply. We are also committed to recruiting more people living with disabilities and strongly encourage them to apply. The employment will be carried out by the central administration of the University of Tübingen.

Do you recognize yourself in the job profile? Then we look forward to receiving your application by September 30, 2021. Please note the position will be filled as soon as an appropriate candidate is found. 

Your application should consist of a single PDF file <lastname_firstname>.pdf containing:

  • A two-page motivational letter, which: 1) explains why you would like to join us and 2) describes the research topics that excite you and that you would like to pursue in your PhD;
  • Your CV, with details of publications and conference participations (if applicable);
  • A copy of your Master’s degree certificate, if you already have one;
  • Unofficial transcripts of all of your university studies (BSc and MSc), as well as a translation into English and explanation of grading system (if needed);
  • Letters of recommendation and/or contact details of 2-3 referees (preferably from different universities);
  • Link to github or enclosed code sample you have written;
  • Optionally, additional documents such as a thesis, published papers, or project portfolios.

You may apply by sending your documents to dan.zhangspam prevention@wsii.uni-tuebingen.de.