I am currently a PhD student at the AI-Lab of the Vrije Universiteit Brussel in Brussels, under the supervision of Prof. Ann Nowé and co-supervision of Prof. Matthew E. Taylor from Washington State University.
My research has spanned multiple topics, but my current focus is on the incorporation of prior knowledge in reinforcement learning. This topic has implications for robotics, transfer learning, learning from demonstration, and reinforcement learning in general. Several publications on these topics are available on this site.
A detailed overview of my activities can be found in my CV.
Contact: firstname.lastname@example.org · Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium
My research has included work on multi-agent reinforcement learning, multi-population evolutionary algorithms and multi-objective reinforcement learning (multi-multi-multi). Currently, I am exploring ways to incorporate abstract heuristic rules, expert demonstrations, expert advice and/or transferred knowledge from previous tasks in reinforcement learning. Read more.
Some code and problem instances we developed and released. The relevant papers are indicated between parentheses.
A short list of selected publications:
Brys, T., Harutyunyan, A., Suay, H.B., Chernova, S., Taylor, M. E., and Nowé, A. (2015).
International Joint Conference on Artificial Intelligence (IJCAI-15).
Brys, T., Harutyunyan, A., Taylor, M. E., and Nowé, A. (2015).
International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS-15).
Brys, T., Nowé, A., Kudenko, D., and Taylor, M. E. (2014).
AAAI Conference on Artificial Intelligence (AAAI-14).
Brys, T., Drugan, M. M., Bosman, P. A. N., De Cock, M., and Nowé, A. (2013).
Genetic and Evolutionary Computation Conference (GECCO-13). Best Paper Award.
An exhaustive list can be found here.