Browsing by Author "Pou Mulet, Bartomeu"
Now showing items 1-5 of 5
Adaptive optics control with multi-agent model-free reinforcement learning Pou Mulet, Bartomeu; Ferreira, Florian; Quiñones Moreno, Eduardo; Gratadour, Damien; Martín Muñoz, Mario (2022-01-14)
Open AccessWe present a novel formulation of closed-loop adaptive optics (AO) control as a multi-agent reinforcement learning (MARL) problem in which the controller is able to learn a non-linear policy and does not need a priori ...
Adaptive optics control with reinforcement learning: first steps Pou Mulet, Bartomeu; Quiñones, Eduardo; Martín Muñoz, Mario (Barcelona Supercomputing Center, 2021-05)
Open AccessWhen planar wavefronts from distant stars traverse the atmosphere, they become distorted due to the atmosphere’s inhomogeneous temperature distribution. Adaptive Optics (AO) is the field in charge of correcting those ...
Denoising wavefront sensor image with deep neural networks Pou Mulet, Bartomeu; Quiñones Moreno, Eduardo; Gratadour, Damien; Martín Muñoz, Mario (International Society for Photo-Optical Instrumentation Engineers (SPIE), 2020)
Open AccessA classical closed-loop adaptive optics system with a Shack-Hartmann wavefront sensor (WFS) relies on a center of gravity approach to process the WFS information and an integrator with gain to produce the commands to a ...
Model-free reinforcement learning with a non-linear reconstructor for closed-loop adaptive optics control with a pyramid wavefront sensor Pou Mulet, Bartomeu; Smith, Jeffrey; Quiñones Moreno, Eduardo; Martín Muñoz, Mario; Gratadour, Damien (International Society for Photo-Optical Instrumentation Engineers (SPIE), 2022)
Open AccessWe present a model-free reinforcement learning (RL) predictive model with a supervised learning non-linear reconstructor for adaptive optics (AO) control with a pyramid wavefront sensor (P-WFS). First, we analyse the ...
Speeding up Reinforcement Learning with Learned Models Pou Mulet, Bartomeu (Universitat Politècnica de Catalunya, 2019-10-16)
Open AccessIn this master thesis, we have tried to solve two of most prominent Reinforcement Learning problems: sparse rewards and sample efficiency. The combination of Model Based Reinforcement Learning, Hindsight Experience Replay ...