My name is Diederik Roijers, and I am an assistant professor in Computational Intelligence at the Vrije Universiteit Amsterdam. I am also a member of the AI lab at the Vrije Universiteit Brussel (VUB). My main research interests are reinforcement learning and planning, and multi-objective decision making. Please check out our tutorials or book for an introduction to multi-objective models and methods for RL and planning, or my publication page for information about our latest research. Other interests of mine are machine learning, evolutionary computing, e-tutoring systems and education.

I did my PhD in the Intelligent Autonomous Systems Group (presently AMLab) at the Informatics Institute at the University of Amsterdam under the supervision of Shimon Whiteson and Frans Oliehoek. Please check out my PhD project for further details. After that I was a postdoctoral researcher in the TERESA project, at the Department of Computer Science of the University of Oxford, and an FWO Postdoctoral Fellow at the Vrije Universiteit Brussel

If you would like to contact me please e-mail me at d(dot)m(dot)roijers(AT)vu{dot}nl or droijers[AT]ai[dot]vub(dot)ac{dot}be. My office is T-333 and located on the third floor at Boelelaan 1081 in Amsterdam (the Netherlands).


January 2018 - Our papers were accepted for publication: "Ordered Preference Elicitation Strategies for Supporting Multi-Objective Decision Making", together with Luisa Zintgraf, Sjoerd Linders, Catholijn Jonker, and Ann Nowé at AAMAS 2018 in Stockholm; and "Bootstrapping LPs in Value Iteration for Multi-Objective and Partially Observable MDPs", together with Erwin Walraven and Matthijs Spaan at ICAPS 2018 in Delft.

November 2017 - Together with Federation University's FLAG group, we created a benchmark suite for multi-objective reinforcement learning called MORL-Glue. Please see our demo paper, video, and code.

November 2017 - Our paper "Reinforcement Learning in POMDPs with Memoryless Options and Option-Observation Initiation Sets" [pdf] was accepted for publication at AAAI 2018.

November 2017 - Our demonstration "Hierarchical Reinforcement Learning for a Robotic Partially Observable Task" [pdf] won the best demo award at BNAIC 2017. Congratulations Denis! ☆

October 2017 - Our student, Tanya Pedersen, won an AMSI Vacation Research Scholarship (VRS) from the Australian Mathematical Sciences Institute (AMSI). Congratulations Tanya! ☆

October 2017 - I am teaching the introduction to RL at the ACAI summer school in Nieuwpoort (Belgium). See this link for details.

May 2017 - Our book "Multi-Objective Decision Making" appeared as an e-book and in print. See this link for details and to purchase the book.

March 2017 - Our paper "Efficient evaluation of influenza mitigation strategies using preventive bandits" was pronounced visionary paper at ALA 2017. ☆

November 2016 - Our student, Hossam Mossalam (UOx) won the best student thesis abstract award for his masters thesis "Multi-Objective Deep Reinforcement Learning with Optimistic Linear Support" [pdf]. Congratulations Hossam! ☆

October 2016 - I started as an FWO Postdoctoral Fellow at the Vrije Universiteit Brussel (VUB).

October 2016 - The tutorial I gave together with Shimon Whiteson on "Multi-objective Planning under Uncertainty" at ICAPS 2016 in June is now available on YouTube.

June 2016 - My FWO Postdoctoral Fellowship proposal titled "Learning with Guarantees in Multi-Objective Decision Problems" was granted. I am very excited to be starting at AI lab at the Vrije Universiteit Brussel from October 1, 2016. ☆

June 2016 - Our paper Monte Carlo Tree Search with Options for General Video Game Playing was accepted for publication at CIG 2016.

June 2016 - I received two peer endorsements for the quality of my reviews for IJCAI.☆

May 24, 2016 - I am now a doctor. [UvA-DARE]☆

April 2016 - My dissertation is now online. My defence will take place on the 24th of May at 10AM in the Agnietenkapel in Amsterdam.

March 2016 - Our paper Balancing Relevance Criteria through Multi-Objective Optimization was accepted for publication at SIGIR 2016.