Dataquitaine is offering a replay of its Data Science Workshop-Webinar, organised by Digital Aquitaine, on 30 June 2020. This seminar was dedicated to the theme “Markovian Decision Process and Learning by Reinforcement” with Emmanuel Hyon.

Problems related to the optimization of a system that evolves over time in a random way and under the effect of agents’ actions are often modelled by Decision Processes or, when the statistics are not known, by reinforcement learning models. Such problems may be shorter stochastic paths, stock or resource management problems.

SPEAKER:

Emmanuel Hyon, Lip6 Lecturer-Researcher and coordinator of the working group on the stochastic optimization aspects of GDR RO, presents the theoretical principles related to the Markovian decision process and learning by reinforcement, illustrated with examples of resource sharing in a cloud.

About DATAQUITAINE Workshops :

As a follow-up to the annual DATAQUITAINE event, the last session of which was held on 12 February 2020, workshops presenting issues and practical cases related to data science will be held at Digital Aquitaine or Data Factory throughout the year.

Translated from Replay Dataquitaine : Processus de Décision Markovien et Apprentissage par Renforcement