Join us for an evening of networking to explore the world of Multi-Agent Reinforcement Learning and discuss current challenges at our AI Happy Hour event! This in-person event will take place on Monday 29 September at 17:30 in the FARI Test and Experience Center (Cantersteen 16, 1000 Brussels).

Yannick Molinghen is a PhD student and a Teaching Assistant at the computer science department of Université Libre de Bruxelles. As a researcher in the Machine Learning Group (MLG), his research revolves around Cooperative Multi-Agent Reinforcement Learning, and more specifically focuses on the problem of reward sparsity. As a teaching assistant, he teaches in many different computer science classes from first year bachelor up to master classes, and supervises multiple master theses in the field of reinforcement learning.

Description of the session: Reinforcement learning can be thought of as a form of training by trial and error, comparable to training an animal to perform new behaviours. A key advantage of this approach is that the trainer - or the algorithm designer - does not need to show the precise sequence of actions that will solve the task but only to give a treat whenever the task has been completed. When we extend RL to the multi-agent setting, where several agents interact and cooperate with each other, this seemingly simple idea becomes considerably more complex. If you give a single treat to a group of animals when one of them does something good, can we expect each of them to understand to whom they owe this treat ? How should we attribute credit to individual agents when success or failure emerges from their joint behaviour? Now, picture an even more complex problem where feedback is delayed or rare as it is the case in many real world problems. How can agents learn effective strategies under such conditions ?

In this talk, we will discuss all of these questions in an approachable way. We will discover together what science can already do in the field of Multi-Agent Reinforcement Learning and which challenges are yet to be solved.

Is this for you? Administrators, industry professionals and AI enthusiasts are invited to join this session.

A balanced mix of learning and relaxation - Attendees will have the opportunity to engage in insightful discussions with Yannick, followed by a casual happy hour with snacks and beverages!

This event is free to attend, but registration is compulsory.This event is organized by FARI - AI for the Common Good Institute.

``Summary

Time: Monday, September 29, 17:30
Place: FARI Test and Experience Center, Cantersteen 16, 1000 Brussels
Speaker: Yannick Molinghen
Snacks & Beverages: Enjoy a casual happy hour
Cost: Free to attend
Target Audience: Administrators, industry professionals, and AI enthusiasts

FARI AI Happy Hour | Multi-Agent Reinforcement Learning