Projects (current)
Projects (past)

Matthijs Spaan


I am an Associate Professor within the Algorithmics group, Delft University of Technology, Delft, The Netherlands.

I held a Marie Curie fellowship in the same group and before I was a senior research scientist at the Institute for Systems and Robotics, Instituto Superior Técnico, Lisbon, Portugal. I have received a PhD degree in Computer Science (2006) and an MSc degree in Artificial Intelligence (2002), both from the University of Amsterdam, The Netherlands.


Faculty position on causal reinforcement learning available (deadline: Mar 31, 2023).Feb 2023
Thiago Dias Simão successfully defended his PhD thesis entitled Safe Online and Offline Reinforcement Learning.Jan 2023
Co-organized a Lorentz workshop on Rigorous Automated Planning.Jun 2022
Co-Chaired the program of AAAI 2022.Feb 2022
Mercury Machine Learning Lab, a collaboration between Booking.com, University of Amsterdam and TU Delft, has been launched. Jun 2021
New paper on safety-constrained reinforcement learning will be published at AAAI-21.Feb 2021
ICAI Lab AI4B.io has been launched.Jan 2021
FET Open project E-pi on epistemic AI has been granted.Dec 2020
Joris Scharpff successfully defended his PhD thesis entitled Collective Decision Making through Self-regulation: Mechanisms and Algorithms for Self-regulation in Decision-Theoretic Planning.Nov 2020
Nils van der Blij successfully defended his PhD thesis entitled DC Distribution Systems: Modeling, Stability, Control & Protection.Sep 2020
IJCAI paper on Structure Learning for Safe Policy Improvement was presented.Aug 2019
Erwin Walraven successfully defended his PhD thesis entitled Planning under Uncertainty in Constrained and Partially Observable Environments.May 2019
Frits de Nijs successfully defended his PhD thesis entitled Resource-constrained Multi-agent Markov Decision Processes.Apr 2019
AAAI paper on safe reinforcement learning was presented.Jan 2019
Horizon 2020 project MEDIATOR on optimal coordination between human driver and autonomous vehicles was granted. PostDoc position on decision making will be available (contact me for details).Jan 2019
Delft AI Research & Education website launched.Dec 2018
JAIR paper on planning for Constrained POMDPs published.Jul 2018
We hosted ICAPS 2018 in Delft.Jun 2018
Elected to the ICAPS Executive Council.Mar 2018
Co-taught a KIVI Big Data Science Master Class on Smart Algorithms for Smart Grids.Nov 2017
Elected to the AAAI Executive Council.Aug 2017
NWO TOP proposal on model learning for sequential decision making granted.Nov 2016
NWO URSES+ proposal on future-proof flexible charging granted. Sep 2016
NWO ESIpose proposal on optimizing flexible energy use in industry granted.Mar 2016

Research overview

My research focuses on decision making under uncertainty for single agents (such as robots) as well as multiagent systems. A major goal of Artificial Intelligence is designing agents: systems that perceive their environment and execute actions. In particular, a fundamental question is how to build intelligent agents. When uncertainty and many agents are involved, this question is particularly challenging and has not yet been answered in a satisfactory way.

Uncertainty manifests itself in various forms when computing plans for agents, in particular in real-world scenarios involving robots. For an agent in isolation, planning under uncertainty in acting and sensing has been studied using decision-theoretic models like Partially Observable Markov Decision Processes (POMDPs). However single-agent, centralized methods do not suffice for large-scale multiagent systems, for which I study multiagent extensions such as the decentralized POMDP (Dec-POMDP) model.

Professional service


You might be interested in