SMILE-IT aims to extend and improve upon the state of the art in multi-agent reinforcement learning and network management in order to implement and validate a generic, stable, and robust multi-agent reinforcement learning framework, capable of (semi-)automatically managing modern networked systems (e.g., telecommunications networks, smart grids, air traffic routing, traffic control) through software.
Powered by Drupal
SMILE-IT
Latest news
UC meeting - Final event
Our fourth and final User-Committee meeting will take place on the 28th of November and it will be part of the IEEE International Forum on Smart Grids for Smart Cities (SG4SC): http://ieeesg4sc.org
We will organise a public session as part of the official program (http://ieeesg4sc.org/program/). After lunch, the event will continue at EnergyVille, where we will showcase demonstrations and have additional presentations and discussions.
User-Committee Meeting
Our third User-Committee meeting took place at the IDLab in Gent. The morning was dedicated to presenting the findings of our researchers, in topics such as traffic control and electric vehicle charging. We also showcased MARLON in a demonstration. The language was used to program multiple Raspberry Pi's, each performing it's own chunk of work of an expensive computation. Reinforcement Learning was used to optimally distribute the chunks of work.
Management Meeting
A Management Meeting took place on January 25. Next to the usual progress report (deliverables and milestones), the MARLON and Electric Vehicle charging responsibles presented an update on their respective works.
Management Meeting
The Smile-IT team gathered together on November 9, for another Management Meeting, at the VUB. The main topics of discussions included the MARLON language, a software language geared towards multi-agent learning algorithms, as well as experiments on using Reinrforcement Learning for thermostatically controlled loads.
Management Meeting
On June 27, the Smile-IT team had yet another Management Meeting. The discussions about smart-charging vehicles sparkled lots of interest in the research team, and early experiment results using Reinforcement Learning to takle this problem were shown. We also examined battery energy management use-cases, as well as vehicle control.