Belgium - Netherlands Reinforcement Learning Workshop 2017,
Thursday 26 January

It is a pleasure to invite you to the next Belgium-Netherlands Reinforcement Learning workshop on January 26 2017 at the Radboud University in Nijmegen. We will try to make this an interdisciplinary meeting, centered on RL, exploring relevant connections with robotics, neuroscience, AI, machine learning and control theory.


Practical information


You can register by sending an email to SNN Annet Wanders,.

Please indicate:
university / company:.....................
phonenumber: ......................
abstract: yes / no
diner (at your own expense): yes / no
special dietary requirements for lunch /diner (if any): ..............


09:30-10:00Welcome with tea coffee
10:00-10:25Talk 1, Hans Ruiz, Donders Nijmegen,
"Smoothing Estimates of Diffusion Processes"
10:25-10:50Talk 2, Katinka van der Kooij, VU Amsterdam,
"Reward-based motor learning: lost in dimensionality"
10:50-11:15Talk 3, Sander Bothe, CWI,
"Spiking Neural Memory for Plausible Reinforcement Learning Models"
11:15-11:45Coffee/ Tea break
11:45-12:10Talk 4, Wouter Kruijne, VU Amsterdam,
"Simultaneous learning of external actions and internal gating to solve multi-item working memory tasks"
12:10-12:35Talk 5, Timothy Verstraeten, VU Brussel,
"Fleet Reinforcement Learning using Dependent Gaussian Processes"
14:00-14:25Talk 6, Roxana Rădulescu, VU Brussel,
"Modelling Congestion Problems in Multi-agent Reinforcement Learning"
14:25-14:50Talk 7, Pablo Hermandez Leal, CWI,
"A Survey of Learning in Multiagent Environments: Dealing with Non-Stationarity"
14:50-15:15Talk 8, Elise van der Pol, University Amsterdam,
"Coordinated Deep Reinforcement Learners for Traffic Light Control"
15:15-15:45Coffee/ Tea break
15:45-16:10Talk 9, Denis Steckelmacher, VU Brussel,
"Extending the Options Model for Reinforcement Learning in POMDPs"
16:10-16:35Talk 10, Ivan Koryakovskiy, TU Delft,
"Evaluation of physical damage associated with action selection strategies in reinforcement learning?"
16:35-17:00Talk 11, Bert Kappen/Dominik Thalmeijer, Donders Nijmegen,
"Action selection in growing state spaces: Control of Network Structure Growth"