6533b870fe1ef96bd12cf1d4
RESEARCH PRODUCT
Warped Gaussian Processes in Remote Sensing Parameter Estimation and Causal Inference
Jordi Munoz-mariGustau Camps-vallsAdrian Perez-suayAnna Mateo-sanchissubject
FOS: Computer and information sciencesComputer Science - Machine LearningHeteroscedasticityRemote sensing applicationComputer scienceComputer Vision and Pattern Recognition (cs.CV)Maximum likelihoodComputer Science - Computer Vision and Pattern Recognition0211 other engineering and technologies02 engineering and technologyBivariate analysis010501 environmental sciences01 natural sciencesMachine Learning (cs.LG)Data modelingsymbols.namesakeElectrical and Electronic EngineeringGaussian process021101 geological & geomatics engineering0105 earth and related environmental sciencesRemote sensingParametric statisticsEstimation theoryHyperspectral imagingGeotechnical Engineering and Engineering GeologyConfidence intervalCausal inferencesymbolsdescription
This letter introduces warped Gaussian process (WGP) regression in remote sensing applications. WGP models output observations as a parametric nonlinear transformation of a GP. The parameters of such a prior model are then learned via standard maximum likelihood. We show the good performance of the proposed model for the estimation of oceanic chlorophyll content from multispectral data, vegetation parameters (chlorophyll, leaf area index, and fractional vegetation cover) from hyperspectral data, and in the detection of the causal direction in a collection of 28 bivariate geoscience and remote sensing causal problems. The model consistently performs better than the standard GP and the more advanced heteroscedastic GP model, both in terms of accuracy and more sensible confidence intervals.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2018-11-01 | IEEE Geoscience and Remote Sensing Letters |