
-
Read more: Parallelisable Memory Recurrent UnitsFlorent De Geeter, Gaspard Lambrechts, Damien Ernst and Guillaume Drion Link: https://hdl.handle.net/2268/339817 Sequence models are important because they are used everywhere: to predict future weather, to take future actions from past observations in reinforcement learning, or even to generate the…
-

Trusting LLMs through attributions: the TreeFinder approach
Read more: Trusting LLMs through attributions: the TreeFinder approachLize Pirenne, Gaspard Lambrechts, Norman Marlier, Maxence de la Brassinne Bonardeaux, Gilles Louppe, Damien Ernst Access to the paper, Access to the code Improving LLM trustworthiness Large Language Models (LLMs) have been adopted to reduce the amount of menial tasks…
-

Les 10 centres d’énergie renouvelable distants les plus prometteurs
Read more: Les 10 centres d’énergie renouvelable distants les plus prometteursAuteurs: Victor Dachet, Damien Ernst and Thibaut Techy Accéder à la présentation ici: https://hdl.handle.net/2268/337189 La consommation d’énergie n’est pas répartie uniformément à travers le monde. Elle se concentre dans certaines régions que nous appelons centres de consommation. En Europe, la…
-

The 10 most promising remote renewable energy hubs
Read more: The 10 most promising remote renewable energy hubsAuthors: Victor Dachet, Damien Ernst and Thibaut Techy Access the presentation: https://hdl.handle.net/2268/337186 Energy consumption is not evenly distributed across the world. It is concentrated in specific regions referred as load centers. In Europe, the Blue Banana is one of the…
-

Gym-TORAX: Breaking Fusion Energy “30 Years Away” Curse with Reinforcement Learning
Read more: Gym-TORAX: Breaking Fusion Energy “30 Years Away” Curse with Reinforcement LearningThere is a saying in energy research: fusion power is always 30 years away. For decades, scientists have inched closer to making it real, but the timeline keeps slipping forward. The reason is not a lack of effort or brilliance,…
-

SecuLEx: a Secure Limit Exchange Market for Dynamic Operating Envelopes
Read more: SecuLEx: a Secure Limit Exchange Market for Dynamic Operating EnvelopesLink to the SecuLEx paper: https://orbi.uliege.be/handle/2268/331735. If you have solar panels, an electric vehicle, or a battery, you are part of a revolution. But this energy revolution is creating hidden problems in the power networks that only a few people…
-

Distributed e-fuel hubs: Concept and case study
Read more: Distributed e-fuel hubs: Concept and case studyAuthors: Samy Mokeddem, Bardhyl Miftari, Victor Dachet, Guillaume Derval, Damien Ernst Access to the paper: https://hdl.handle.net/2268/327044 Sectors such as steel, cement, aviation, and shipping are among the most challenging to decarbonize, due to their reliance on high-temperature heat, which is…
-

Two PhD positions in Operations Research
Read more: Two PhD positions in Operations ResearchWe are looking for two PhD students who will work on a topic at the intersection of Operations Research and Energy Systems: energy systems models’ transparency. By using techniques from multiple OR and ML fields (sensitivity analysis, parametric optimization, machine…
-

New paper on the techno-economic performance of Small Modular Reactors!
Read more: New paper on the techno-economic performance of Small Modular Reactors!Authors: Antoine Larbanois, Bardhyl Miftari, Antoine Mouchamps, Ayyildiz Kerem Enes, Vincent Schryvers, Guillaume Derval, Damien Ernst Access to the paper: https://hdl.handle.net/2268/331891 This work is the result of a great collaboration between the University of Liège and the SCK CEN, the…
-

A Theoretical Justification for Asymmetric Actor-Critic Algorithms
Read more: A Theoretical Justification for Asymmetric Actor-Critic AlgorithmsGaspard Lambrechts, Damien Ernst and Aditya Mahajan Access the paper: https://hdl.handle.net/2268/326874 Our paper entitled “A Theoretical Justification for Asymmetric Actor-Critic Algorithms” was accepted at ICML 2025! Never heard of asymmetric actor-critic algorithms? Yet, many successful Reinforcement Learning (RL) applications use…

