Data augmentation through expert-guided symmetry detection to improve performance in offline reinforcement learning

G Angelotti, N Drougard, CPC Chanel - arxiv preprint arxiv:2112.09943, 2021 - arxiv.org
Offline estimation of the dynamical model of a Markov Decision Process (MDP) is a non-
trivial task that greatly depends on the data available in the learning phase. Sometimes the …

Data-Efficient Offline Reinforcement Learning with Approximate Symmetries

G Angelotti, N Drougard, CPC Chanel - International Conference on …, 2023 - Springer
Abstract The performance of Offline Reinforcement Learning (ORL) models in Markov
Decision Processes (MDPs) is heavily contingent upon the quality and diversity of the …

[PDF][PDF] The partially observable brain: An exploratory study on the use of partially observable Markov decision processes as a general framework for brain-computer …

JJT TRESOLS - 2024 - theses.fr
Although POMDP shows promise in being integrated into BCI pipelines as a unifying,
generic, flexible and extensible decision framework, it presents the glaring shortcoming of …

[PDF][PDF] Crop Optimization in Space with Machine learning & Offline computation of Strategies–COSMOS

N DROUGARD–ISAE-SUPAERO… - isae-supaero.fr
Advances in space exploration require human beings to be in space for the long term.
Indeed, permanent settlement on other planets is now very much on the agenda, as are long …