I am hiring!

I am looking for several, fully funded, PhD students and postdocs to work on my ERC starting grant “INFLUENCE”, see more information below.

Please contact me if you are interested in my research, or if you have students that would make strong candidates!


I am afraid that I cannot offer internships. Please do not email me requesting internships; I am afraid I cannot respond to these requests.

Vacancies in ERC Project ‘INFLUENCE’

I have been awarded a prestigious ERC Starting Grant (€1.48M) for the project INLFUENCE: INFLUEnce-based decision-making in uNCertain Environments. The project will employ a team research team of 3 students and 2 postdocs led by myself. I’m looking for most people to start in the period May – Sept. 2018.

Project description

Decision-theoretic sequential decision making (SDM) is concerned with endowing an intelligent agent with the capability to choose actions that optimize task performance. SDM techniques have the potential to revolutionize many aspects of society, and recent successes, e.g., agents that play Atari game and beat a world champion in the game of Go, have sparked renewed interest in this field.

However, despite these successes, fundamental problems of scalability prevent these methods from addressing other problems with hundreds or thousands of state variables. To overcome this barrier, INFLUENCE will develop a new class of influence-based SDM methods that overcome scalability issues for such problems by using novel ways of abstraction.

For instance, when we think of controlling traffic lights in an entire city, an intersection’s local problem is manageable, but the influence that the rest of the network exerts on it is complex. The key idea is that by using (deep) machine learning methods, we can learn sufficiently accurate representations of such influence to facilitate near-optimal decisions.

This project will aim to develop novel decision making methods and demonstrate their scalability on two simulated challenge domains:

  • traffic lights control in an entire city, and
  • robotic order picking in a large-scale autonomous warehouse.

(But also other domains can be considered.)

General requirements for all candidates

Since this is a project which aims to create a true synergy between its team members, it is important that all team members have excellent communication skills and are willing to contribute to a team effort. Demonstrable experience on a team coding project is highly desirable.


PhD student in Machine Learning of Influence Descriptions in Complex Systems

This position will focus on learning compact descriptions of ‘influence’, i.e., how a sub-problem is affected over time, by building on state-of-the-art (deep) machine learning techniques. This requires excellent math skills, and background in machine learning is required, and demonstrable with state-of-the-art deep learning frameworks, such as tensorflow, is highly desirable. Additionally, the candidate would benefit from some experience in C/C++ in order to aid in development of simulation frameworks.

Available: Four years from summer 2018, with some flexibility.

PhD student in Influence-based Sequential Decision Making

This position will focus online planning methods (such as extensions of Monte Carlo tree search methods) making use of learned influence descriptions, and developing self-improving simulators. The candidate will need very strong algorithmic skills, and have some background in reinforcement learning or planning under uncertainty (e.g., familiar with Markov decision processes). Additionally he or she requires excellent coding skills and should ideally be proficient in C++.

Available: Four years from summer 2018, with some flexibility.

PhD student in Influence-based Reinforcement Learning and Exploration

This position will focus on exploiting the potential of compact influence representations in the context of reinforcement learning. Excellent mathematical skills are required and the candidate should have had some exposure to reinforcement learning algorithms. Good coding skills are required.

Available:  Four years from end of 2018, flexible.

Postdoc in High Performance Planning and Learning Algorithms

This position focuses on high-performance simulation-based planning and learning, and the candidate will play a leading role in the joint development efforts. As such, the candidate should have ample experience in team development projects as well as implementing complex high-performance algorithms in C++. The candidate should also have top-tier publications in the general area of reinforcement learning and decision making under uncertainty.

Available: Three year position, from summer 2018, with some flexibility.

Postdoc in Influence-based Abstraction, Learning and Coordination

This position will focus on theoretical aspects of decision making under uncertainty and multiagent learning and coordination. The candidate should have an excellent track record in the general area of machine learning, reinforcement learning or planning under uncertainty, as evidenced by publications at top-tier venues, and should have demonstrable experience in theoretical analysis of algorithms (e.g., proving approximation or sample bounds).

Available: Three year position, from fall 2018, flexible.


You will need to apply via the HR department of TU Delft.

Once the vacancies go live, I will post the links on this page and write a post, so please subscribe to my feed if you are interested.

Contact information

Dr. Frans Oliehoek