Proceedings of the Fifth European workshop on Reinforcement Learning

Abstract

The Fifth European Workshop on Reinforcement Learning (EWRL-5) has gathered a wide variety of researchers interested in many different topics in reinforcement learning (RL). First of all, several papers describe RL algorithms for solving POMDPs. In this category, there is a paper by Hartley and Wyatt describing using Hidden Markov Models (HMMs), where state splitting is used to learn the right model. A paper by Dutech and Scherrer describes using contextual information for handling POMDPs, and a paper by Bakker describes using LSTM recurrent networks for solving POMDPs.