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

TL;DR

This paper presents an active learning approach using Gaussian processes to efficiently create surrogate models of complex, costly computer codes, with applications demonstrated in remote sensing.

Contribution

It introduces the Active Multi-Output Gaussian Process Emulator (AMOGAPE), a novel sequential method that adaptively constructs accurate, tractable emulators for multi-output computer models.

Findings

AMOGAPE achieves high accuracy with fewer samples.

The method effectively captures low-density and fluctuating regions.

Promising results demonstrated on remote sensing transfer code.

Abstract

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, model tractability, as well as compact and expressive simulated datasets. In order to achieve this, the proposed Active Multi-Output Gaussian Process Emulator (AMOGAPE) combines the predictive capacity of Gaussian Processes (GPs) with the design of an acquisition function that favors sampling in low density and fluctuating regions of the approximation functions. Comparing different acquisition functions, we illustrate the promising performance of the method for the construction of emulators with toy examples, as well as for a widely used remote sensing transfer code.