Search Results for author: Gustau Camps-Valls

Found 79 papers, 17 papers with code

Recovering Latent Confounders from High-dimensional Proxy Variables

1 code implementation21 Mar 2024 Nathan Mankovich, Homer Durand, Emiliano Diaz, Gherardo Varando, Gustau Camps-Valls

Detecting latent confounders from proxy variables is an essential problem in causal effect estimation.

Causal Graph Neural Networks for Wildfire Danger Prediction

no code implementations13 Mar 2024 Shan Zhao, Ioannis Prapas, Ilektra Karasante, Zhitong Xiong, Ioannis Papoutsis, Gustau Camps-Valls, Xiao Xiang Zhu

In that direction, we propose integrating causality with Graph Neural Networks (GNNs) that explicitly model the causal mechanism among complex variables via graph learning.

Decision Making Graph Learning

Improving generalisation via anchor multivariate analysis

no code implementations4 Mar 2024 Homer Durand, Gherardo Varando, Nathan Mankovich, Gustau Camps-Valls

We introduce a causal regularisation extension to anchor regression (AR) for improved out-of-distribution (OOD) generalisation.

Causal Inference

Causal hybrid modeling with double machine learning

no code implementations20 Feb 2024 Kai-Hendrik Cohrs, Gherardo Varando, Nuno Carvalhais, Markus Reichstein, Gustau Camps-Valls

Hybrid modeling integrates machine learning with scientific knowledge to enhance interpretability, generalization, and adherence to natural laws.

Causal Inference

Fun with Flags: Robust Principal Directions via Flag Manifolds

1 code implementation8 Jan 2024 Nathan Mankovich, Gustau Camps-Valls, Tolga Birdal

In this work, we present a unifying formalism for PCA and its variants, and introduce a framework based on the flags of linear subspaces, ie a hierarchy of nested linear subspaces of increasing dimension, which not only allows for a common implementation but also yields novel variants, not explored previously.

Dimensionality Reduction

Assessing the Causal Impact of Humanitarian Aid on Food Security

no code implementations17 Oct 2023 Jordi Cerdà-Bautista, José María Tárraga, Vasileios Sitokonstantinou, Gustau Camps-Valls

In the face of climate change-induced droughts, vulnerable regions encounter severe threats to food security, demanding urgent humanitarian assistance.

Causal Inference Humanitarian

TeleViT: Teleconnection-driven Transformers Improve Subseasonal to Seasonal Wildfire Forecasting

1 code implementation19 Jun 2023 Ioannis Prapas, Nikolaos Ioannis Bountos, Spyros Kondylatos, Dimitrios Michail, Gustau Camps-Valls, Ioannis Papoutsis

To achieve such accurate long-term forecasts at a global scale, it is crucial to employ models that account for the Earth system's inherent spatio-temporal interactions, such as memory effects and teleconnections.


Discovering Causal Relations and Equations from Data

no code implementations21 May 2023 Gustau Camps-Valls, Andreas Gerhardus, Urmi Ninad, Gherardo Varando, Georg Martius, Emili Balaguer-Ballester, Ricardo Vinuesa, Emiliano Diaz, Laure Zanna, Jakob Runge

Discovering equations, laws and principles that are invariant, robust and causal explanations of the world has been fundamental in physical sciences throughout the centuries.


Understanding cirrus clouds using explainable machine learning

1 code implementation3 May 2023 Kai Jeggle, David Neubauer, Gustau Camps-Valls, Ulrike Lohmann

Their dependencies on meteorological and aerosol conditions are among the largest uncertainties in global climate models.

The Kernelized Taylor Diagram

1 code implementation18 May 2022 Kristoffer Wickstrøm, J. Emmanuel Johnson, Sigurd Løkse, Gustau Camps-Valls, Karl Øyvind Mikalsen, Michael Kampffmeyer, Robert Jenssen

Our proposed kernelized Taylor diagram is capable of visualizing similarities between populations with minimal assumptions of the data distributions.

Data Visualization

Unsupervised Anomaly and Change Detection with Multivariate Gaussianization

no code implementations12 Apr 2022 José A. Padrón-Hidalgo, Valero Laparra, Gustau Camps-Valls

Identifying low probability events in data/images is a challenging problem given the high-dimensionality of the data, especially when no (or little) information about the anomaly is available a priori.

Anomaly Detection Change Detection

Inference over radiative transfer models using variational and expectation maximization methods

1 code implementation7 Apr 2022 Daniel Heestermans Svendsen, Daniel Hernández-Lobato, Luca Martino, Valero Laparra, Alvaro Moreno, Gustau Camps-Valls

Radiative transfer models (RTMs) encode the energy transfer through the atmosphere, and are used to model and understand the Earth system, as well as to estimate the parameters that describe the status of the Earth from satellite observations by inverse modeling.

Earth Observation

Graph Embedding via High Dimensional Model Representation for Hyperspectral Images

no code implementations29 Nov 2021 Gulsen Taskin, Gustau Camps-Valls

Learning the manifold structure of remote sensing images is of paramount relevance for modeling and understanding processes, as well as to encapsulate the high dimensionality in a reduced set of informative features for subsequent classification, regression, or unmixing.

Graph Embedding Vocal Bursts Intensity Prediction

Integrating Domain Knowledge in Data-driven Earth Observation with Process Convolutions

no code implementations16 Apr 2021 Daniel Heestermans Svendsen, Maria Piles, Jordi Muñoz-Marí, David Luengo, Luca Martino, Gustau Camps-Valls

We specifically propose the use of a class of GP convolution models called latent force models (LFMs) for EO time series modelling, analysis and understanding.

Earth Observation Time Series +1

Deep Gaussian Processes for Biogeophysical Parameter Retrieval and Model Inversion

no code implementations16 Apr 2021 Daniel Heestermans Svendsen, Pablo Morales-Alvarez, Ana Belen Ruescas, Rafael Molina, Gustau Camps-Valls

Currently, different approximations exist: a direct, yet costly, inversion of radiative transfer models (RTMs); the statistical inversion with in situ data that often results in problems with extrapolation outside the study area; and the most widely adopted hybrid modeling by which statistical models, mostly nonlinear and non-parametric machine learning algorithms, are applied to invert RTM simulations.

Earth Observation Gaussian Processes +1

Semisupervised Manifold Alignment of Multimodal Remote Sensing Images

no code implementations15 Apr 2021 Devis Tuia, Michele Volpi, Maxime Trolliet, Gustau Camps-Valls

We introduce a method for manifold alignment of different modalities (or domains) of remote sensing images.

Image Classification

Towards a Collective Agenda on AI for Earth Science Data Analysis

1 code implementation11 Apr 2021 Devis Tuia, Ribana Roscher, Jan Dirk Wegner, Nathan Jacobs, Xiao Xiang Zhu, Gustau Camps-Valls

In the last years we have witnessed the fields of geosciences and remote sensing and artificial intelligence to become closer.


Learning Structures in Earth Observation Data with Gaussian Processes

no code implementations22 Dec 2020 Fernando Mateo, Jordi Munoz-Mari, Valero Laparra, Jochem Verrelst, Gustau Camps-Valls

Gaussian Processes (GPs) has experienced tremendous success in geoscience in general and for bio-geophysical parameter retrieval in the last years.

Earth Observation Gaussian Processes +1

Gap Filling of Biophysical Parameter Time Series with Multi-Output Gaussian Processes

no code implementations11 Dec 2020 Anna Mateo-Sanchis, Jordi Munoz-Mari, Manuel Campos-Taberner, Javier Garcia-Haro, Gustau Camps-Valls

In this work we evaluate multi-output (MO) Gaussian Process (GP) models based on the linear model of coregionalization (LMC) for estimation of biophysical parameter variables under a gap filling setup.

Gaussian Processes Time Series +1

A deep network approach to multitemporal cloud detection

no code implementations9 Dec 2020 Devis Tuia, Benjamin Kellenberger, Adrian Pérez-Suay, Gustau Camps-Valls

With a single model, we are able to outline clouds along all year and during day and night with high accuracy.

Cloud Detection Time Series +1

Spatial noise-aware temperature retrieval from infrared sounder data

no code implementations9 Dec 2020 David Malmgren-Hansen, Valero Laparra, Allan Aasbjerg Nielsen, Gustau Camps-Valls

We compare Principal Component Analysis (PCA) and Minimum Noise Fraction (MNF) for dimensionality reduction, and study the compactness and information content of the extracted features.

Dimensionality Reduction regression +1

Convolutional Neural Networks for Multispectral Image Cloud Masking

no code implementations9 Dec 2020 Gonzalo Mateo-García, Luis Gómez-Chova, Gustau Camps-Valls

Convolutional neural networks (CNN) have proven to be state of the art methods for many image classification tasks and their use is rapidly increasing in remote sensing problems.

BIG-bench Machine Learning Classification +2

Consistent regression of biophysical parameters with kernel methods

no code implementations9 Dec 2020 Emiliano Díaz, Adrián Pérez-Suay, Valero Laparra, Gustau Camps-Valls

This paper introduces a novel statistical regression framework that allows the incorporation of consistency constraints.


Disentangling Derivatives, Uncertainty and Error in Gaussian Process Models

1 code implementation9 Dec 2020 Juan Emmanuel Johnson, Valero Laparra, Gustau Camps-Valls

Gaussian Processes (GPs) are a class of kernel methods that have shown to be very useful in geoscience applications.

Earth Observation Gaussian Processes +1

Pattern Recognition Scheme for Large-Scale Cloud Detection over Landmarks

no code implementations8 Dec 2020 Adrián Pérez-Suay, Julia Amorós-López, Luis Gómez-Chova, Jordi Muñoz-Marí, Dieter Just, Gustau Camps-Valls

Landmark recognition and matching is a critical step in many Image Navigation and Registration (INR) models for geostationary satellite services, as well as to maintain the geometric quality assessment (GQA) in the instrument data processing chain of Earth observation satellites.

Cloud Detection Earth Observation +1

Nonlinear Cook distance for Anomalous Change Detection

no code implementations8 Dec 2020 José A. Padrón Hidalgo, Adrián Pérez-Suay, Fatih Nar, Gustau Camps-Valls

In this work we propose a method to find anomalous changes in remote sensing images based on the chronochrome approach.

Change Detection

Randomized RX for target detection

no code implementations8 Dec 2020 Fatih Nar, Adrián Pérez-Suay, José Antonio Padrón, Gustau Camps-Valls

This work tackles the target detection problem through the well-known global RX method.

Retrieval of Case 2 Water Quality Parameters with Machine Learning

1 code implementation8 Dec 2020 Ana B. Ruescas, Gonzalo Mateo-Garcia, Gustau Camps-Valls, Martin Hieronymi

Water quality parameters are derived applying several machine learning regression methods on the Case2eXtreme dataset (C2X).

BIG-bench Machine Learning regression +1

Sparsity-driven Digital Terrain Model Extraction

no code implementations7 Dec 2020 Fatih Nar, Erdal Yilmaz, Gustau Camps-Valls

We here introduce an automatic Digital Terrain Model (DTM) extraction method.

Model extraction

Efficient Nonlinear RX Anomaly Detectors

no code implementations7 Dec 2020 José A. Padrón Hidalgo, Adrián Pérez-Suay, Fatih Nar, Gustau Camps-Valls

In this letter, we propose two families of techniques to improve the efficiency of the standard kernel Reed-Xiaoli (RX) method for anomaly detection by approximating the kernel function with either {\em data-independent} random Fourier features or {\em data-dependent} basis with the Nystr\"om approach.

Anomaly Detection

Spectral band selection for vegetation properties retrieval using Gaussian processes regression

no code implementations7 Dec 2020 Jochem Verrelst, Juan Pablo Rivera, Anatoly Gitelson, Jesus Delegido, José Moreno, Gustau Camps-Valls

GPR-BAT allows (1) to identify the most informative bands in relating spectral data to a biophysical variable, and (2) to find the least number of bands that preserve optimized accurate predictions.

GPR regression +1

Gradient-based Automatic Look-Up Table Generator for Atmospheric Radiative Transfer Models

no code implementations7 Dec 2020 Jorge Vicent, Luis Alonso, Luca Martino, Neus Sabater, Jochem Verrelst, Gustau Camps-Valls

Our results indicate that, when compared to a pseudo-random homogeneous distribution of the LUT nodes, GALGA reduces (1) the LUT size by $\sim$75\% and (2) the maximum interpolation relative errors by 0. 5\% It is concluded that automatic LUT design might benefit from the methodology proposed in GALGA to reduce computation time and interpolation errors.

Earth Observation

Causal Inference in Geoscience and Remote Sensing from Observational Data

no code implementations7 Dec 2020 Adrián Pérez-Suay, Gustau Camps-Valls

Establishing causal relations between random variables from observational data is perhaps the most important challenge in today's \blue{science}.

Causal Inference regression

Randomized kernels for large scale Earth observation applications

no code implementations7 Dec 2020 Adrián Pérez-Suay, Julia Amorós-López, Luis Gómez-Chova, Valero Laparra, Jordi Muñoz-Marí, Gustau Camps-Valls

Dealing with land cover classification of the new image sources has also turned to be a complex problem requiring large amount of memory and processing time.

Classification Earth Observation +5

Emulation as an Accurate Alternative to Interpolation in Sampling Radiative Transfer Codes

no code implementations7 Dec 2020 Jorge Vicent, Jochem Verrelst, Juan Pablo Rivera-Caicedo, Neus Sabater, Jordi Muñoz-Marí, Gustau Camps-Valls, José Moreno

Computationally expensive Radiative Transfer Models (RTMs) are widely used} to realistically reproduce the light interaction with the Earth surface and atmosphere.

GPR regression

Active Learning Methods for Efficient Hybrid Biophysical Variable Retrieval

no code implementations7 Dec 2020 ochem Verrelst, Sara Dethier, Juan Pablo Rivera, Jordi Muñoz-Marí, Gustau Camps-Valls, José Moreno

Kernel-based machine learning regression algorithms (MLRAs) are potentially powerful methods for being implemented into operational biophysical variable retrieval schemes.

Active Learning Retrieval

Nonlinear Distribution Regression for Remote Sensing Applications

no code implementations7 Dec 2020 Jose E. Adsuara, Adrián Pérez-Suay, Jordi Muñoz-Marí, Anna Mateo-Sanchis, Maria Piles, Gustau Camps-Valls

When the target variable is available at a resolution that matches the remote sensing observations, standard algorithms such as neural networks, random forests or Gaussian processes are readily available to relate the two.

Gaussian Processes Multiple Instance Learning +1

Causal Inference in Geosciences with Kernel Sensitivity Maps

no code implementations7 Dec 2020 Adrián Pérez-Suay, Gustau Camps-Valls

Establishing causal relations between random variables from observational data is perhaps the most important challenge in today's Science.

Causal Inference regression

Multi-temporal and multi-source remote sensing image classification by nonlinear relative normalization

no code implementations7 Dec 2020 Devis Tuia, Diego Marcos, Gustau Camps-Valls

Remote sensing image classification exploiting multiple sensors is a very challenging problem: data from different modalities are affected by spectral distortions and mis-alignments of all kinds, and this hampers re-using models built for one image to be used successfully in other scenes.

General Classification Image Classification +1

Understanding Climate Impacts on Vegetation with Gaussian Processes in Granger Causality

no code implementations6 Dec 2020 Miguel Morata-Dolz, Diego Bueso, Maria Piles, Gustau Camps-Valls

Global warming is leading to unprecedented changes in our planet, with great societal, economical and environmental implications, especially with the growing demand of biofuels and food.

Gaussian Processes

Fusing Optical and SAR time series for LAI gap filling with multioutput Gaussian processes

no code implementations5 Dec 2020 Luca Pipia, Jordi Muñoz-Marí, Eatidal Amin, Santiago Belda, Gustau Camps-Valls, Jochem Verrelst

The availability of satellite optical information is often hampered by the natural presence of clouds, which can be problematic for many applications.

Gaussian Processes Time Series +1

Information Theory in Density Destructors

2 code implementations2 Dec 2020 J. Emmanuel Johnson, Valero Laparra, Gustau Camps-Valls, Raul Santos-Rodríguez, Jesús Malo

Density destructors are differentiable and invertible transforms that map multivariate PDFs of arbitrary structure (low entropy) into non-structured PDFs (maximum entropy).

Gaussianizing the Earth: Multidimensional Information Measures for Earth Data Analysis

3 code implementations13 Oct 2020 J. Emmanuel Johnson, Valero Laparra, Maria Piles, Gustau Camps-Valls

Information theory is an excellent framework for analyzing Earth system data because it allows us to characterize uncertainty and redundancy, and is universally interpretable.

Density Estimation

Information Theory Measures via Multidimensional Gaussianization

4 code implementations8 Oct 2020 Valero Laparra, J. Emmanuel Johnson, Gustau Camps-Valls, Raul Santos-Rodríguez, Jesus Malo

Information theory is an outstanding framework to measure uncertainty, dependence and relevance in data and systems.

Density Estimation

A Perspective on Gaussian Processes for Earth Observation

no code implementations2 Jul 2020 Gustau Camps-Valls, Dino Sejdinovic, Jakob Runge, Markus Reichstein

Earth observation (EO) by airborne and satellite remote sensing and in-situ observations play a fundamental role in monitoring our planet.

Causal Inference Earth Observation +2

Accounting for Input Noise in Gaussian Process Parameter Retrieval

1 code implementation20 May 2020 J. Emmanuel Johnson, Valero Laparra, Gustau Camps-Valls

In this letter, we demonstrate how one can account for input noise estimates using a GP model formulation which propagates the error terms using the derivative of the predictive mean function.

Earth Observation Gaussian Processes +1

Active emulation of computer codes with Gaussian processes -- Application to remote sensing

no code implementations13 Dec 2019 Daniel Heestermans Svendsen, Luca Martino, Gustau Camps-Valls

Many fields of science and engineering rely on running simulations with complex and computationally expensive models to understand the involved processes in the system of interest.

Active Learning Gaussian Processes

Kernel Dependence Regularizers and Gaussian Processes with Applications to Algorithmic Fairness

no code implementations11 Nov 2019 Zhu Li, Adrian Perez-Suay, Gustau Camps-Valls, Dino Sejdinovic

We present a regularization approach to this problem that trades off predictive accuracy of the learned models (with respect to biased labels) for the fairness in terms of statistical parity, i. e. independence of the decisions from the sensitive covariates.

Crime Prediction Ethics +2

Joint Gaussian Processes for Biophysical Parameter Retrieval

no code implementations14 Nov 2017 Daniel Heestermans Svendsen, Luca Martino, Manuel Campos-Taberner, Francisco Javier García-Haro, Gustau Camps-Valls

Radiative transfer models (RTMs) represent mathematically the physical laws which govern the phenomena in remote sensing applications (forward models).

Gaussian Processes regression +1

Fair Kernel Learning

no code implementations16 Oct 2017 Adrián Pérez-Suay, Valero Laparra, Gonzalo Mateo-García, Jordi Muñoz-Marí, Luis Gómez-Chova, Gustau Camps-Valls

It has been shown that not including sensitive features that bias fairness, such as gender or race, is not enough to mitigate the discrimination when other related features are included.

BIG-bench Machine Learning Dimensionality Reduction +2

Remote Sensing Image Classification with Large Scale Gaussian Processes

no code implementations2 Oct 2017 Pablo Morales-Alvarez, Adrian Perez-Suay, Rafael Molina, Gustau Camps-Valls

Current remote sensing image classification problems have to deal with an unprecedented amount of heterogeneous and complex data sources.

Classification Cloud Detection +4

The Recycling Gibbs Sampler for Efficient Learning

no code implementations21 Nov 2016 Luca Martino, Victor Elvira, Gustau Camps-Valls

The key point for the successful application of the Gibbs sampler is the ability to draw efficiently samples from the full-conditional probability density functions.

Bayesian Inference Computational Efficiency +1

Optimized Kernel Entropy Components

no code implementations9 Mar 2016 Emma Izquierdo-Verdiguier, Valero Laparra, Robert Jenssen, Luis Gómez-Chova, Gustau Camps-Valls

Results show that 1) OKECA returns projections with more expressive power than KECA, 2) the most successful rule for estimating the kernel parameter is based on maximum likelihood, and 3) OKECA is more robust to the selection of the length-scale parameter in kernel density estimation.

Density Estimation

Dimensionality Reduction via Regression in Hyperspectral Imagery

no code implementations31 Jan 2016 Valero Laparra, Jesus Malo, Gustau Camps-Valls

DRR identifies the nonlinear features through multivariate regression to ensure the reduction in redundancy between he PCA coefficients, the reduction of the variance of the scores, and the reduction in the reconstruction error.

Dimensionality Reduction Land Cover Classification +1

Unsupervised Deep Feature Extraction for Remote Sensing Image Classification

no code implementations25 Nov 2015 Adriana Romero, Carlo Gatta, Gustau Camps-Valls

This paper introduces the use of single layer and deep convolutional networks for remote sensing data analysis.

Classification General Classification +4

Kernel Manifold Alignment

1 code implementation9 Apr 2015 Devis Tuia, Gustau Camps-Valls

We introduce a kernel method for manifold alignment (KEMA) and domain adaptation that can match an arbitrary number of data sources without needing corresponding pairs, just few labeled examples in all domains.

Computational Efficiency Domain Adaptation +1

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