Stochmod EURO PhD School
Reinforcement learning applied to operations research: 01.06.-08.06.2021.
In this PhD school we want to build a bridge between Markov Decision Processes (MDP) and Reinforcement Learning (RL) and give OR student studying or using MDP’s the possibility to learn about RL techniques. RL can be seen as a collection of techniques or heuristics to solve large-scale MDP’s. It is a part of computer science and developed mostly independently from MDPs. RL is part of machine learning, together with unsupervised and supervised learning. After the current boom of supervised learning, with deep learning as its prime example, RL is expected to be the next big thing. For example, it is the technology that enables autonomous vehicles to think ahead, and it is equally important as deep learning which is used for image recognition. In this school we want to invite a few speakers from RL to teach us the methods and typical applications of RL. Then we will apply this to some problems that come from OR. We will take time to implement some of the algorithms to make sure the methods are well understood and to test them on some relatively easy instances of typical OR problems.
More information on the Stochmod EURO PhD School is comming soon.