Search Results for author: Lucas Drumetz

Found 23 papers, 12 papers with code

Physics Informed and Data Driven Simulation of Underwater Images via Residual Learning

1 code implementation7 Feb 2024 Tanmoy Mondal, Ricardo Mendoza, Lucas Drumetz

We propose a deep learning-based architecture to automatically simulate the underwater effects where only a dehazing-like image formation equation is known to the network, and the additional degradation due to the other unknown factors if inferred in a data-driven way.

Image Dehazing

Time-changed normalizing flows for accurate SDE modeling

no code implementations22 Dec 2023 Naoufal El Bekri, Lucas Drumetz, Franck Vermet

The generative paradigm has become increasingly important in machine learning and deep learning models.

Gaussian Processes Time Series

MultiHU-TD: Multifeature Hyperspectral Unmixing Based on Tensor Decomposition

2 code implementations5 Oct 2023 Mohamad Jouni, Mauro Dalla Mura, Lucas Drumetz, Pierre Comon

Matrix models become insufficient when the hyperspectral image (HSI) is represented as a high-order tensor with additional features in a multimodal, multifeature framework.

Hyperspectral Unmixing Tensor Decomposition

Neural Koopman prior for data assimilation

1 code implementation11 Sep 2023 Anthony Frion, Lucas Drumetz, Mauro Dalla Mura, Guillaume Tochon, Abdeldjalil Aïssa El Bey

With the increasing availability of large scale datasets, computational power and tools like automatic differentiation and expressive neural network architectures, sequential data are now often treated in a data-driven way, with a dynamical model trained from the observation data.

Self-Supervised Learning Time Series

Learning Sentinel-2 reflectance dynamics for data-driven assimilation and forecasting

1 code implementation5 May 2023 Anthony Frion, Lucas Drumetz, Guillaume Tochon, Mauro Dalla Mura, Abdeldjalil Aïssa El Bey

Over the last few years, massive amounts of satellite multispectral and hyperspectral images covering the Earth's surface have been made publicly available for scientific purpose, for example through the European Copernicus project.

Self-Supervised Learning Time Series

Leveraging Neural Koopman Operators to Learn Continuous Representations of Dynamical Systems from Scarce Data

no code implementations13 Mar 2023 Anthony Frion, Lucas Drumetz, Mauro Dalla Mura, Guillaume Tochon, Abdeldjalil Aissa El Bey

Over the last few years, several works have proposed deep learning architectures to learn dynamical systems from observation data with no or little knowledge of the underlying physics.

Sliced-Wasserstein on Symmetric Positive Definite Matrices for M/EEG Signals

2 code implementations10 Mar 2023 Clément Bonet, Benoît Malézieux, Alain Rakotomamonjy, Lucas Drumetz, Thomas Moreau, Matthieu Kowalski, Nicolas Courty

When dealing with electro or magnetoencephalography records, many supervised prediction tasks are solved by working with covariance matrices to summarize the signals.

Brain Computer Interface Computational Efficiency +5

Disambiguation of One-Shot Visual Classification Tasks: A Simplex-Based Approach

1 code implementation16 Jan 2023 Yassir Bendou, Lucas Drumetz, Vincent Gripon, Giulia Lioi, Bastien Pasdeloup

Then, we introduce a downstream classifier meant to exploit the presence of multiple objects to improve the performance of few-shot classification, in the case of extreme settings where only one shot is given for its class.

Hyperbolic Sliced-Wasserstein via Geodesic and Horospherical Projections

1 code implementation18 Nov 2022 Clément Bonet, Laetitia Chapel, Lucas Drumetz, Nicolas Courty

It has been shown beneficial for many types of data which present an underlying hierarchical structure to be embedded in hyperbolic spaces.

Image Classification

Geometry-preserving Lie Group Integrators For Differential Equations On The Manifold Of Symmetric Positive Definite Matrices

no code implementations17 Oct 2022 Lucas Drumetz, Alexandre Reiffers-Masson, Naoufal El Bekri, Franck Vermet

The application of Euclidean methods to integrate differential equations lying on such objects does not respect the geometry of the manifold, which can cause many numerical issues.

Numerical Integration

Active Few-Shot Classification: a New Paradigm for Data-Scarce Learning Settings

no code implementations23 Sep 2022 Aymane Abdali, Vincent Gripon, Lucas Drumetz, Bartosz Boguslawski

We consider a novel formulation of the problem of Active Few-Shot Classification (AFSC) where the objective is to classify a small, initially unlabeled, dataset given a very restrained labeling budget.

Active Learning

Turning Normalizing Flows into Monge Maps with Geodesic Gaussian Preserving Flows

1 code implementation22 Sep 2022 Guillaume Morel, Lucas Drumetz, Simon Benaïchouche, Nicolas Courty, François Rousseau

Normalizing Flows (NF) are powerful likelihood-based generative models that are able to trade off between expressivity and tractability to model complex densities.

Preserving Fine-Grain Feature Information in Classification via Entropic Regularization

1 code implementation7 Aug 2022 Raphael Baena, Lucas Drumetz, Vincent Gripon

In this paper we are interested in the problem of solving fine-grain classification or regression, using a model trained on coarse-grain labels only.

Classification Few-Shot Learning +3

Spherical Sliced-Wasserstein

1 code implementation17 Jun 2022 Clément Bonet, Paul Berg, Nicolas Courty, François Septier, Lucas Drumetz, Minh-Tan Pham

Many variants of the Wasserstein distance have been introduced to reduce its original computational burden.

Density Estimation Variational Inference

Preventing Manifold Intrusion with Locality: Local Mixup

1 code implementation12 Jan 2022 Raphael Baena, Lucas Drumetz, Vincent Gripon

Mixup is a data-dependent regularization technique that consists in linearly interpolating input samples and associated outputs.

Image Classification

Efficient Gradient Flows in Sliced-Wasserstein Space

1 code implementation21 Oct 2021 Clément Bonet, Nicolas Courty, François Septier, Lucas Drumetz

However, it requires solving a nested optimization problem at each iteration, and is known for its computational challenges, especially in high dimension.

Bayesian Inference Image Generation

Subspace Detours Meet Gromov-Wasserstein

no code implementations21 Oct 2021 Clément Bonet, Nicolas Courty, François Septier, Lucas Drumetz

In the context of optimal transport methods, the subspace detour approach was recently presented by Muzellec and Cuturi (2019).

Graphs as Tools to Improve Deep Learning Methods

no code implementations8 Oct 2021 Carlos Lassance, Myriam Bontonou, Mounia Hamidouche, Bastien Pasdeloup, Lucas Drumetz, Vincent Gripon

This chapter is composed of four main parts: tools for visualizing intermediate layers in a DNN, denoising data representations, optimizing graph objective functions and regularizing the learning process.

Denoising

Learning stochastic dynamical systems with neural networks mimicking the Euler-Maruyama scheme

no code implementations18 May 2021 Noura Dridi, Lucas Drumetz, Ronan Fablet

They are notable for the ability to include a deterministic component of the system and a stochastic one to represent random unknown factors.

Inferring Graph Signal Translations as Invariant Transformations for Classification Tasks

no code implementations18 Feb 2021 Raphael Baena, Lucas Drumetz, Vincent Gripon

The field of Graph Signal Processing (GSP) has proposed tools to generalize harmonic analysis to complex domains represented through graphs.

Classification General Classification

Learning Sentinel-2 Spectral Dynamics for Long-Run Predictions Using Residual Neural Networks

no code implementations17 Nov 2020 Joaquim Estopinan, Guillaume Tochon, Lucas Drumetz

Making the most of multispectral image time-series is a promising but still relatively under-explored research direction because of the complexity of jointly analyzing spatial, spectral and temporal information.

Time Series Time Series Analysis

Cannot find the paper you are looking for? You can Submit a new open access paper.