Výsledky vyhledávání - "Guen, Vincent Le"

Report

Complementing Brightness Constancy with Deep Networks for Optical Flow Prediction

Autor: Guen, Vincent Le, Rambour, Clément, Thome, Nicolas

State-of-the-art methods for optical flow estimation rely on deep learning, which require complex sequential training schemes to reach optimal performances on real-world data. In this work, we introduce the COMBO deep network that explicitly exploits

Externí odkaz: http://arxiv.org/abs/2207.03790

Zobrazit plný text záznamu

Report

Deep learning for spatio-temporal forecasting -- application to solar energy

Autor: Guen, Vincent Le

This thesis tackles the subject of spatio-temporal forecasting with deep learning. The motivating application at Electricity de France (EDF) is short-term solar energy forecasting with fisheye images. We explore two main research directions for impro

Externí odkaz: http://arxiv.org/abs/2205.03571

Zobrazit plný text záznamu

Report

Deep Time Series Forecasting with Shape and Temporal Criteria

Autor: Guen, Vincent Le, Thome, Nicolas

This paper addresses the problem of multi-step time series forecasting for non-stationary signals that can present sudden changes. Current state-of-the-art deep learning forecasting methods, often trained with variants of the MSE, lack the ability to

Externí odkaz: http://arxiv.org/abs/2104.04610

Zobrazit plný text záznamu

Report

Probabilistic Time Series Forecasting with Structured Shape and Temporal Diversity

Autor: Guen, Vincent Le, Thome, Nicolas

Probabilistic forecasting consists in predicting a distribution of possible future outcomes. In this paper, we address this problem for non-stationary time series, which is very challenging yet crucially important. We introduce the STRIPE model for r

Externí odkaz: http://arxiv.org/abs/2010.07349

Zobrazit plný text záznamu

Report

Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting

Autor: Yin, Yuan, Guen, Vincent Le, Dona, Jérémie, de Bézenac, Emmanuel, Ayed, Ibrahim, Thome, Nicolas, Gallinari, Patrick

Publikováno v: J. Stat. Mech. (2021) 124012

Forecasting complex dynamical phenomena in settings where only partial knowledge of their dynamics is available is a prevalent problem across various scientific fields. While purely data-driven approaches are arguably insufficient in this context, st

Externí odkaz: http://arxiv.org/abs/2010.04456

Zobrazit plný text záznamu

Report

Disentangling Physical Dynamics from Unknown Factors for Unsupervised Video Prediction

Autor: Guen, Vincent Le, Thome, Nicolas

Leveraging physical knowledge described by partial differential equations (PDEs) is an appealing way to improve unsupervised video prediction methods. Since physics is too restrictive for describing the full visual content of generic videos, we intro

Externí odkaz: http://arxiv.org/abs/2003.01460

Zobrazit plný text záznamu

Report

Shape and Time Distortion Loss for Training Deep Time Series Forecasting Models

Autor: Guen, Vincent Le, Thome, Nicolas

This paper addresses the problem of time series forecasting for non-stationary signals and multiple future steps prediction. To handle this challenging task, we introduce DILATE (DIstortion Loss including shApe and TimE), a new objective function for

Externí odkaz: http://arxiv.org/abs/1909.09020

Zobrazit plný text záznamu

A divide-and-conquer approach for genomic prediction in rubber tree using machine learning

Autor: Aono, Alexandre Hild, Francisco, Felipe Roberto, Souza, Livia Moura, Gonçalves, Paulo de Souza, Scaloppi, Erivaldo J., Guen, Vincent Le, Fritsche-Neto, Roberto, Gorjanc, Gregor, Quiles, Marcos Gonçalves, de Souza, Anete Pereira

Publikováno v: Aono, A H, Francisco, F R, Souza, L M, Gonçalves, P D S, Scaloppi, E J, Guen, V L, Fritsche-Neto, R, Gorjanc, G, Quiles, M G & de Souza, A P 2022 ' A divide-and-conquer approach for genomic prediction in rubber tree using machine learning ' bioRxiv . https://doi.org/10.1101/2022.03.30.486381

Rubber tree (Hevea brasiliensis) is the main feedstock for commercial rubber; however, its long vegetative cycle has hindered the development of more productive varieties via breeding programs. With the availability of H. brasiliensis genomic data, s

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______3094::0bcdb807d07c6e4e00ba02e5c68ec444
https://hdl.handle.net/20.500.11820/8c8ff985-0de3-4853-ac3e-e5229fbfb144

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání