Search results for "Computer Science - Machine Learning"

showing 10 items of 155 documents

Retrieval of Case 2 Water Quality Parameters with Machine Learning

2018

Water quality parameters are derived applying several machine learning regression methods on the Case2eXtreme dataset (C2X). The used data are based on Hydrolight in-water radiative transfer simulations at Sentinel-3 OLCI wavebands, and the application is done exclusively for absorbing waters with high concentrations of coloured dissolved organic matter (CDOM). The regression approaches are: regularized linear, random forest, Kernel ridge, Gaussian process and support vector regressors. The validation is made with and an independent simulation dataset. A comparison with the OLCI Neural Network Swarm (ONSS) is made as well. The best approached is applied to a sample scene and compared with t…

FOS: Computer and information sciencesComputer Science - Machine Learning010504 meteorology & atmospheric sciences0211 other engineering and technologiesFOS: Physical sciences02 engineering and technologyMachine learningcomputer.software_genre01 natural sciencesData modelingMachine Learning (cs.LG)Physics - Geophysicssymbols.namesakeRadiative transferGaussian process021101 geological & geomatics engineering0105 earth and related environmental sciencesMathematicsArtificial neural networkbusiness.industry6. Clean waterRandom forestGeophysics (physics.geo-ph)Support vector machineColored dissolved organic matterKernel (statistics)Physics - Data Analysis Statistics and ProbabilitysymbolsArtificial intelligencebusinesscomputerData Analysis Statistics and Probability (physics.data-an)
researchProduct

Retrieval of coloured dissolved organic matter with machine learning methods

2017

The coloured dissolved organic matter (CDOM) concentration is the standard measure of humic substance in natural waters. CDOM measurements by remote sensing is calculated using the absorption coefficient (a) at a certain wavelength (e.g. 440nm). This paper presents a comparison of four machine learning methods for the retrieval of CDOM from remote sensing signals: regularized linear regression (RLR), random forest (RF), kernel ridge regression (KRR) and Gaussian process regression (GPR). Results are compared with the established polynomial regression algorithms. RLR is revealed as the simplest and most efficient method, followed closely by its nonlinear counterpart KRR.

FOS: Computer and information sciencesComputer Science - Machine Learning010504 meteorology & atmospheric sciences0211 other engineering and technologiesFOS: Physical sciences02 engineering and technologyMachine learningcomputer.software_genre01 natural sciencesMachine Learning (cs.LG)Physics - GeophysicsKrigingDissolved organic carbonLinear regression021101 geological & geomatics engineering0105 earth and related environmental sciencesMathematicsPolynomial regressionbusiness.industry6. Clean waterGeophysics (physics.geo-ph)Random forestNonlinear systemColored dissolved organic matterKernel (statistics)Artificial intelligencebusinesscomputer
researchProduct

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

2018

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. In particular, we focus on LAI and fAPAR over rice areas. We show how this problem cannot be solved with standard single-output (SO) GP models, and how the proposed MO-GP models are able to successfully predict these variables even in high missing data regimes, by implicitly performing an across-domain information transfer.

FOS: Computer and information sciencesComputer Science - Machine Learning010504 meteorology & atmospheric sciences0211 other engineering and technologiesFOS: Physical sciencesMachine Learning (stat.ML)02 engineering and technology01 natural sciencesQuantitative Biology - Quantitative MethodsMachine Learning (cs.LG)Data modelingsymbols.namesakeStatistics - Machine LearningApplied mathematicsTime seriesGaussian processQuantitative Methods (q-bio.QM)021101 geological & geomatics engineering0105 earth and related environmental sciencesMathematicsSeries (mathematics)Linear modelProbability and statisticsMissing dataFOS: Biological sciencesPhysics - Data Analysis Statistics and ProbabilitysymbolsFocus (optics)Data Analysis Statistics and Probability (physics.data-an)
researchProduct

Disentangling Derivatives, Uncertainty and Error in Gaussian Process Models

2020

Gaussian Processes (GPs) are a class of kernel methods that have shown to be very useful in geoscience applications. They are widely used because they are simple, flexible and provide very accurate estimates for nonlinear problems, especially in parameter retrieval. An addition to a predictive mean function, GPs come equipped with a useful property: the predictive variance function which provides confidence intervals for the predictions. The GP formulation usually assumes that there is no input noise in the training and testing points, only in the observations. However, this is often not the case in Earth observation problems where an accurate assessment of the instrument error is usually a…

FOS: Computer and information sciencesComputer Science - Machine Learning010504 meteorology & atmospheric sciencesComputer science0211 other engineering and technologiesMachine Learning (stat.ML)02 engineering and technology01 natural sciencesMachine Learning (cs.LG)symbols.namesakeStatistics - Machine LearningGaussian process021101 geological & geomatics engineering0105 earth and related environmental sciencesVariance functionPropagation of uncertaintyVariance (accounting)Function (mathematics)Confidence intervalNonlinear systemNoiseKernel method13. Climate actionKernel (statistics)symbolsAlgorithmIGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium
researchProduct

A Deep Network Approach to Multitemporal Cloud Detection

2018

We present a deep learning model with temporal memory to detect clouds in image time series acquired by the Seviri imager mounted on the Meteosat Second Generation (MSG) satellite. The model provides pixel-level cloud maps with related confidence and propagates information in time via a recurrent neural network structure. With a single model, we are able to outline clouds along all year and during day and night with high accuracy.

FOS: Computer and information sciencesComputer Science - Machine Learning010504 meteorology & atmospheric sciencesComputer scienceFeature extraction0211 other engineering and technologiesCloud detectionFOS: Physical sciencesCloud computing02 engineering and technologyCloud detection01 natural sciencesMachine Learning (cs.LG)Laboratory of Geo-information Science and Remote SensingLaboratorium voor Geo-informatiekunde en Remote Sensing021101 geological & geomatics engineering0105 earth and related environmental sciencesRemote sensingbusiness.industrySeviriDeep learningDeep learningPE&RCPhysics - Atmospheric and Oceanic PhysicsRecurrent neural networkRecurrent neural networksAtmospheric and Oceanic Physics (physics.ao-ph)Convolutional neural networksSatelliteArtificial intelligencebusinessNetwork approachIGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium
researchProduct

Cloud detection machine learning algorithms for PROBA-V

2020

This paper presents the development and implementation of a cloud detection algorithm for Proba-V. Accurate and automatic detection of clouds in satellite scenes is a key issue for a wide range of remote sensing applications. With no accurate cloud masking, undetected clouds are one of the most significant sources of error in both sea and land cover biophysical parameter retrieval. The objective of the algorithms presented in this paper is to detect clouds accurately providing a cloud flag per pixel. For this purpose, the method exploits the information of Proba-V using statistical machine learning techniques to identify the clouds present in Proba-V products. The effectiveness of the propo…

FOS: Computer and information sciencesComputer Science - Machine Learning010504 meteorology & atmospheric sciencesComputer scienceRemote sensing applicationFeature extraction0211 other engineering and technologiesFOS: Physical sciencesCloud computing02 engineering and technologyLand coverMachine learningcomputer.software_genre01 natural sciencesMachine Learning (cs.LG)Astrophysics::Galaxy Astrophysics021101 geological & geomatics engineering0105 earth and related environmental sciencesPixelbusiness.industrySupport vector machinePhysics - Atmospheric and Oceanic PhysicsAtmospheric and Oceanic Physics (physics.ao-ph)Artificial intelligencebusinesscomputerAlgorithm2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)
researchProduct

Symmetry meets AI

2021

We explore whether Neural Networks (NNs) can {\it discover} the presence of symmetries as they learn to perform a task. For this, we train hundreds of NNs on a {\it decoy task} based on well-controlled Physics templates, where no information on symmetry is provided. We use the output from the last hidden layer of all these NNs, projected to fewer dimensions, as the input for a symmetry classification task, and show that information on symmetry had indeed been identified by the original NN without guidance. As an interdisciplinary application of this procedure, we identify the presence and level of symmetry in artistic paintings from different styles such as those of Picasso, Pollock and Van…

FOS: Computer and information sciencesComputer Science - Machine Learning0303 health sciencesTheoretical computer scienceArtificial neural networkComputer Vision and Pattern Recognition (cs.CV)PhysicsQC1-999Computer Science - Computer Vision and Pattern RecognitionFOS: Physical sciencesGeneral Physics and Astronomy01 natural sciencesMachine Learning (cs.LG)Task (project management)High Energy Physics - Phenomenology03 medical and health sciencesHigh Energy Physics - Phenomenology (hep-ph)0103 physical sciencesHomogeneous spacePICASSOHidden layerSymmetry (geometry)010306 general physics030304 developmental biologySciPost Physics
researchProduct

Learning User's Confidence for Active Learning

2013

In this paper, we study the applicability of active learning in operative scenarios: more particularly, we consider the well-known contradiction between the active learning heuristics, which rank the pixels according to their uncertainty, and the user's confidence in labeling, which is related to both the homogeneity of the pixel context and user's knowledge of the scene. We propose a filtering scheme based on a classifier that learns the confidence of the user in labeling, thus minimizing the queries where the user would not be able to provide a class for the pixel. The capacity of a model to learn the user's confidence is studied in detail, also showing the effect of resolution is such a …

FOS: Computer and information sciencesComputer Science - Machine LearningActive learning (machine learning)Computer scienceComputer Vision and Pattern Recognition (cs.CV)SVM0211 other engineering and technologiesComputer Science - Computer Vision and Pattern RecognitionContext (language use)02 engineering and technologyMachine learningcomputer.software_genreTask (project management)Machine Learning (cs.LG)Classifier (linguistics)0202 electrical engineering electronic engineering information engineeringFOS: Electrical engineering electronic engineering information engineeringbad statesElectrical and Electronic Engineeringphotointerpretationuser's confidence021101 geological & geomatics engineeringActive learning (AL)Pixelbusiness.industryRank (computer programming)Image and Video Processing (eess.IV)very high resolution (VHR) imagery020206 networking & telecommunicationsElectrical Engineering and Systems Science - Image and Video ProcessingClass (biology)General Earth and Planetary SciencesArtificial intelligenceHeuristicsbusinesscomputerIEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
researchProduct

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

2020

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. Nevertheless, the high cost involved hamper reliable and exhaustive simulations. Very often such codes incorporate heuristics that ironically make them less tractable and transparent. This paper introduces an active learning methodology for adaptively constructing surrogate models, i.e. emulators, of such costly computer codes in a multi-output setting. The proposed technique is sequential and adaptive, and is based on the optimization of a suitable acquisition function. It aims to achieve accurate approximations…

FOS: Computer and information sciencesComputer Science - Machine LearningActive learningActive learning (machine learning)Computer sciencemedia_common.quotation_subjectMachine Learning (stat.ML)Radiative transfer model02 engineering and technology01 natural sciencesMachine Learning (cs.LG)symbols.namesakeArtificial IntelligenceStatistics - Machine Learning0103 physical sciences0202 electrical engineering electronic engineering information engineeringCode (cryptography)Emulation010306 general physicsFunction (engineering)Gaussian processGaussian process emulatorGaussian processRemote sensingmedia_commonEmulationbusiness.industrySampling (statistics)Remote sensingSignal ProcessingGlobal Positioning Systemsymbols020201 artificial intelligence & image processingComputer codeComputer Vision and Pattern RecognitionbusinessHeuristicsSoftwareDesign of experimentsPattern Recognition
researchProduct

The Tsetlin Machine -- A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic

2018

Although simple individually, artificial neurons provide state-of-the-art performance when interconnected in deep networks. Arguably, the Tsetlin Automaton is an even simpler and more versatile learning mechanism, capable of solving the multi-armed bandit problem. Merely by means of a single integer as memory, it learns the optimal action in stochastic environments through increment and decrement operations. In this paper, we introduce the Tsetlin Machine, which solves complex pattern recognition problems with propositional formulas, composed by a collective of Tsetlin Automata. To eliminate the longstanding problem of vanishing signal-to-noise ratio, the Tsetlin Machine orchestrates the au…

FOS: Computer and information sciencesComputer Science - Machine LearningArtificial Intelligence (cs.AI)Computer Science - Artificial IntelligenceComputer Vision and Pattern Recognition (cs.CV)Computer Science - Computer Vision and Pattern RecognitionMachine Learning (cs.LG)
researchProduct