963 citations
- Olivier Sigaud2 profiles23 · h 35
- M. Chetouani13 · h 44
- Nicolas Bredèche2 profiles13 · h 24
- P. Oudeyer11 · h 56
- Cédric Colas9 · h 20
- S. Doncieux9 · h 30
- Benoît Girard7 · h 25
- C. Pelachaud2 profiles7 · h 52
- Nicolas Obin2 profiles7 · h 18
- Alexandre Coninx6 · h 13
- Catherine Pelachaud6 · h 3
- Laure Soulier2 profiles6 · h 3
- Sorbonne UniversitéFR141 papers
- Centre National de la Recherche ScientifiqueFR89 papers
- Université Pierre-et-Marie-CurieFR22 papers
- Université Paris 1 Panthéon-SorbonneFR9 papers
- Université Paris-SaclayFR8 papers
- Criteo (France)FR7 papers
- Institut national de recherche en sciences et technologies du numériqueFR7 papers
- Télécom ParisFR7 papers
- InsermFR6 papers
- Institut de Recherche et Coordination Acoustique MusiqueFR6 papers
- Sciences et Technologies de la Musique et du SonFR6 papers
- Centre d'études et de recherche en informatique et communicationsFR5 papers
46 papers · 1 filter
Guided Flow Policy: Learning from High-Value Actions in Offline Reinforcement Learning
Franki Nguimatsia Tiofack, Théotime Le Hellard, Fabian Schramm +2
Offline reinforcement learning often relies on behavior regularization that enforces policies to remain close to the dataset distribution. However, such approaches fail to distingu…
ENMA: Tokenwise Autoregression for Generative Neural PDE Operators
Armand Kassaï Koupaï, Lise Le Boudec, Louis Serrano +1
Solving time-dependent parametric partial differential equations (PDEs) remains a fundamental challenge for neural solvers, particularly when generalizing across a wide range of ph…
Learning to explore when mistakes are not allowed
Charly Pecqueux-Guézénec, Stéphane Doncieux, Nicolas Perrin-Gilbert
Goal-Conditioned Reinforcement Learning (GCRL) provides a versatile framework for developing unified controllers capable of handling wide ranges of tasks, exploring environments, a…
Reinforcement Learning for Aligning Large Language Models Agents with Interactive Environments: Quantifying and Mitigating Prompt Overfitting
Mohamed Salim Aissi, Clement Romac, Thomas Carta +5
Reinforcement learning (RL) is a promising approach for aligning large language models (LLMs) knowledge with sequential decision-making tasks. However, few studies have thoroughly…
Supra-Laplacian Encoding for Transformer on Dynamic Graphs
Yannis Karmim, Marc Lafon, Raphael Fournier S'niehotta +1
Fully connected Graph Transformers (GT) have rapidly become prominent in the static graph community as an alternative to Message-Passing models, which suffer from a lack of express…
AFU: Actor-Free critic Updates in off-policy RL for continuous control
Nicolas Perrin-Gilbert
This paper presents AFU, an off-policy deep RL algorithm addressing in a new way the challenging "max-Q problem" in Q-learning for continuous action spaces, with a solution based o…