PhD - Machine Learning and Game Theory for Multi-Agent Systems
- Robert-Bosch-Campus 1, 71272 Renningen, Germany
- Legal Entity: Robert Bosch GmbH
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!
More and more machine learning (ML) agents interact with other ML agents and/or humans: e.g. in autonomous driving, smart cities or decentralized power generation. Therefore, learning models of such interactions and optimizing ML agents’ decisions in such interactions becomes increasingly important. This poses highly non-trivial challenges in terms of general learning-/game-theoretic understanding and tractable implementation.
In this PhD thesis, you will work with us to push the frontier in research on model learning and decision making for such multi-agent interactions:
- You develop novel ML model classes with good inductive biases, algorithms, and/or mathematical theory (prove theorems etc.) that combine ML with game-theoretic principles and reasoning (game theory is the mathematical theory of strategic interacting agents).
- On the machine learning side, you will use, e.g. implicit layers, graph neural nets and/or generative adversarial networks (GANs). On the game-theoretic side, you build on, e.g. Markov games and/or mechanism design. Relevant areas also include (inverse) multi-agent reinforcement learning, interpretable ML and adversarially-robust ML.
- Code and evaluate your algorithms on relevant data sets and tasks.
- You publish papers at top-tier conferences (NIPS, ICML, AAAI, UAI, etc.) and/or journals (JMLR etc.), develop a substantial understanding of the relevant existing work and keep close contact with the academic community.
- Be part of a leading AI industry research lab, the Bosch Center for Artificial Intelligence (BCAI). Participate in academic interactions.
- You have no industry project duties – you perform exclusively academic research with excellence.
- Education: excellent master degree in mathematics, computer science, physics or similar
- Personality: good communication and team work skills
- Working Practice: structured, independently and inquisitive
- Experience and Knowledge: Very good math skills, good coding skills, familiar with machine learning. A background in game theory/multi-agent systems and prior experience in scientific writing would be ideally
- Enthusiasm: A strong passion for doing top-level research and a genuine interest in multi-agent/collective systems and how they can be improved using machine learning
- Languages: very good in English (written and spoken)
The following papers are examples of our research in this direction:
- Geiger, P., & Straehle, C.-N. (2020). Multiagent trajectory models via game theory and implicit layer-based learning. ArXiv Preprint ArXiv: 2008.07303. (https://arxiv.org/pdf/2008.07303.pdf)
- Geiger, P., Besserve, M., Winkelmann, J., Proissl, C., & Schölkopf, B. (2019). Coordinating users of shared facilities based on data-driven assistants and game-theoretic analysis. In Proceedings of the 35th Conference on Uncertainty in Artificial Intelligence (UAI). (https://arxiv.org/pdf/1803.06247.pdf)
Please submit all relevant documents (incl. curriculum vitae, cover letter, certificates). Please be aware that by submitting your application, you agree with us sharing your documents with the academic supervisor of this project.
The final PhD topic is subject to your university. Duration: 3 years
Apply now in just 3 minutes!
Are you interested in working from home, or part-time? Please don't hesitate to ask us.
Need support during your application?
Kevin Heiner (Human Resources)
+49 711 811 12223
Need further information about the job?
Philipp Geiger (Functional Department)
+49 711 811 92277