Abstract—In this study, we evaluated the predictive performance of various deep neural network approaches for spatial data. Recent advances in neural networks have led to a surge in deep learning methods and applications. Many of these methods were developed implicitly for independent data, and it is not trivial to incorporate spatial correlation into them. However, several strategies have been proposed to adapt neural networks to account for spatial correlation. Using both simulated and real-world datasets, we compared spatial methods developed from fully connected deep neural networks, analyzing the impact of different spatial adaptations. Our findings aim to inform future research directions in this evolving field.
 
Keywords—deep learning, Kriging, neural network, prediction, spatial statistics


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).