6533b81ffe1ef96bd1278e17

RESEARCH PRODUCT

Automatic Emulation by Adaptive Relevance Vector Machines

Gustau Camps-vallsJorge VicentLuca Martino

subject

EmulationComputer scienceFlatness (systems theory)0211 other engineering and technologies02 engineering and technologyFunction (mathematics)Relevance vector machineSampling (signal processing)Lookup table0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingRelevance (information retrieval)Algorithm021101 geological & geomatics engineeringInterpolation

description

This paper introduces an automatic methodology to construct emulators for costly radiative transfer models (RTMs). The proposed method is sequential and adaptive, and it is based on the notion of the acquisition function by which instead of optimizing the unknown RTM underlying function we propose to achieve accurate approximations. The proposed methodology combines the interpolation capabilities of a modified Relevance Vector Machine (RVM) with the accurate design of an acquisition function that favors sampling in low density regions and flatness of the interpolation function. The proposed Relevance Vector Machine Automatic Emulator (RAE) is illustrated in toy examples and for the construction of an optimal look-up-table for atmospheric correction based on MODTRAN5.

https://doi.org/10.1007/978-3-319-59126-1_37