PhD – Meta Learning from Few Tasks with Hybrid Models

  • Robert-Bosch-Campus 1, 71272 Renningen, Germany
  • Full-time
  • Legal Entity: Robert Bosch GmbH

Company Description

Do you want beneficial technologies being shaped by your ideas? Whether in the areas of mobility solutions, consumer goods, industrial technology or energy and building technology – with us, you will have the chance to improve quality of life all across the globe. Welcome to Bosch.

The Robert Bosch GmbH is looking forward to your application!

Job Description

Single-task learning aims to generalize over a sample set collected for a specific learning task (e.g. 10-class object classification from images). Contrarily, meta-learning assumes few examples available for progressively many tasks and aims to learn the regularities that govern the occurrence of these tasks. Meta-learning aims to infer a distribution on tasks, so that a sensible learner can be sampled for a new task for which only a handful of observations are available. Hence, they directly target extrapolation settings, which call for a proper uncertainty treatment. This PhD position is about development of novel meta-learning methods that benefit maximally from well-calibrated uncertainties provided from Bayesian task predictors. During your PhD you will be part of the active research team at the Bosch Center for Artificial Intelligence (BCAI, 

  • Help shape the future: You invent novel ways to integrate Bayesian neural networks into the meta-learning setup.
  • Observe, and think ahead: You conduct prototypical algorithm implementations and benchmarking on real-world data sets.
  • Take responsibility: You publish in top-tier conferences (ICML, NIPS, ICLR, AISTATS, etc.) and journals (JMLR, PAMI, etc.).
  • Experience cooperation: You screen literature and build up close contact with the academic community.
  • Networked communication: You participate in academic interactions within the BCAI research team.


  • Personality: enthusiastic for science, excited to tackle theoretical challenges, team player
  • Working Practice: perform exclusively academic research with excellence (i.e. no industry project duties), gradually develop scientific independence until graduation, make significant contributions to science
  • Experience and Knowledge: general knowledge of machine learning, excellent theoretical skills proven by top course grades, prior experience on scientific writing is a plus, a background on Gaussian processes or Bayesian inference is preferable
  • Enthusiasm: intrinsic motivation for top-level research and scientific independence
  • Languages: fluent in English (written and spoken)
  • Education: master degree in machine learning, mathematics, statistics, physics, computer science, or related fields with excellent grades

Additional Information

The final PhD topic is subject to your university.
Duration: 3 years                                                                                 

Need support during your application?
Kevin Heiner (Human Resources)
+49 711 811 12223  

Need further information about the job?
Melih Kandemir (Functional Department)
+49 711 811-52789

Videos To Watch

Privacy Policy