Nearest-Neighbor Neural Networks for Geostatistics
This provides a more flexible spatial prediction method for geostatistics applications, though it appears incremental as it builds on existing neural network approaches with domain-specific adaptations.
The authors tackled the limitations of traditional kriging methods in spatial prediction by proposing a Nearest-Neighbor Neural Network (4N) process that integrates deep learning into geostatistics, showing it outperforms state-of-the-art methods on simulated non-Gaussian data and applying it to a massive forestry dataset.
Kriging is the predominant method used for spatial prediction, but relies on the assumption that predictions are linear combinations of the observations. Kriging often also relies on additional assumptions such as normality and stationarity. We propose a more flexible spatial prediction method based on the Nearest-Neighbor Neural Network (4N) process that embeds deep learning into a geostatistical model. We show that the 4N process is a valid stochastic process and propose a series of new ways to construct features to be used as inputs to the deep learning model based on neighboring information. Our model framework outperforms some existing state-of-art geostatistical modelling methods for simulated non-Gaussian data and is applied to a massive forestry dataset.