RL@Alpes 2023

(Last updated: 27.02.2022)

RL@Alpes is a Reinforcement Learning (RL) Meeting at Grenoble. It provides a venue to meet and connect with scientists from the larger Grenoble area who research or apply RL.

Where:  Inria Grenoble (655 Avenue de l’Europe, 38334 Montbonnot), Room: Grand Amphithéâtre
(Inria can be reached with bus line C1 going to the bus stop INRIA)

When:  March 1st, 2023 (Wednesday)

Who:  Scientists, industrial partners, and everyone that is interested or involved in the research and application of RL.

Program

The talks are live-streamed on YouTube on March 1st via: https://www.youtube.com/watch?v=87mF1aN2G_M

All times are CET (Central European Time).

  09:10

Welcome
  09:20

Sihem Amer-Yahia
(CNRS, UGA, LIG)

 DRL for Data Exploration
  09:40

Nassim Bouarour
(LIG)

 DRL for Recommendations
  10:00 Mohamed Sana
(CEA-Leti)		 Distributed Multi-agent RL for Radio Resource Orchestration in B5G Networks
 

10:20

Break

  
  10:40 Marc Dymetman
(Naverlabs Europe)

Germán Kruszewski
(Naverlabs Europe)

Do we need RL to Align Language Models to Human Preferences?
  11:20

Bruno Gaujal
(Inria)

 RL in a Birth and Death Process
  11:40

Victor Boone
(Inria)

 The Regret of Exploration
  12:00

Lunch

 (Lunch is self-covered. Please see below for a list of possible restaurants.)
  13:20

Chris Reinke
(Inria)

 Successor Feature Representations: Transfer RL for Social Robotics
  13:40

Anand Balou
(Inria)

 Variational Meta-RL for Social Robotics
  14:00

Julien Perez
(Naverlabs Europe)

 Reusable Skill Learning with Safety Guarantees for Robotic Manipulation
  14:20

Rémy Rigo-Mariani
(CNRS, G2Elab)

 RL for Applications in Power and Energy Systems
  14:40

Break

  
  15:00

Darko Dakulic
(Naverlabs Europe)

Arnaud Sors
(Naverlabs Europe)

 RL for Combinatorial Optimization
  15:40

Andrea Mattioni
(GIPSA, UGA)

 DRL Control of Tokamak Safety Factor
  16:00

Paul Aubin
(Inria)

 Battery Energy Storage Optimization on the Energy Market
  16:20

Break

  
  16:40

Alena Shilova
(Inria)

 Continuous-Time RL
  17:00
(1h)

Keynote Talk
Martha White
(University of Alberta)

 Leveraging Offline Data to Calibrate Online RL Agents
(see below for more information)
  20:00

(After Hours)

 For those interested in more socializing, we will meet at the brasserie La Table Ronde in Grenoble for an after-hours drink.

Just come and find us there.

Keynote Talk by Prof. Martha White

Title:  Leveraging Offline Data to Calibrate Online Reinforcement Learning Agents

Abstract: 
Reinforcement Learning provides a general formalism to address control problems, but has yet to be widely used in the real-world. We have been investigating using reinforcement learning to optimize drinking water treatment, directly on the physical system, and have naturally run into many technical challenges. In this talk, I will discuss one such important question for real-world deployment: selecting hyperparameters offline, for a continual learning agent. Before a learning agent can be deployed, its hyperparameters need to be specified. I will outline one potential solution, under the setting where an offline batch of data is available from a previous controller in the system, and discuss key open challenges.

Bio:
Martha White is an Associate Professor of Computing Science at the University of Alberta and a PI of Amii–the Alberta Machine Intelligence Institute–which is one of the top machine learning centres in the world. She holds a Canada CIFAR AI Chair and received IEEE’s “AIs 10 to Watch: The Future of AI” award in 2020. She has authored more than 50 papers in top journals and conferences. Martha is an associate editor for JMLR and TPAMI, and has served as co-program chair for ICLR and area chair for many conferences in AI and ML, including ICML, NeurIPS, AAAI and IJCAI. 

Breaks and Lunch

We provide coffee and cookies for the breaks.
Lunch and After-Hours are self-covered. For Lunch, there are restaurants close by:

For the speakers, a table is reserved for lunch at “La Bonne Pâte”.

Registration

If you want to join in person at Inria please register via: https://sondages.inria.fr/index.php/622289
Participation is free.

You can also join online via our YouTube stream: https://www.youtube.com/watch?v=87mF1aN2G_M
No registration is required for it.

Organizers

The Robotlearn team of Inria Grenoble, represented by Chris Reinke (PostDoc) and Xavier Alameda Pineda (Teamleader).
Our team is researching Deep RL methods for social robotics as part of the European H2020 SPRING project. We are focusing on Transfer and Meta RL to adapt the robot quickly to different reward functions and user requirements.

Contact

chris.reinke@inria.fr


(RL@Alpes Logo adapted from Freepik)

Comments are closed.