Model Evaluation for Geospatial Problems

Abstract

Geospatial problems often involve spatial autocorrelation and covariate shift, which violate the independent, identically distributed assumption underlying standard cross-validation. In this work, we establish a theoretical criterion for unbiased cross-validation, introduce a preliminary categorization framework to guide practitioners in choosing suitable cross-validation strategies for geospatial problems, reconcile conflicting recommendations on best practices, and develop a novel, straightforward method with both theoretical guarantees and empirical success.

Cite

Text

Wang et al. "Model Evaluation for Geospatial Problems." NeurIPS 2023 Workshops: CompSust, 2023.

Markdown

[Wang et al. "Model Evaluation for Geospatial Problems." NeurIPS 2023 Workshops: CompSust, 2023.](https://mlanthology.org/neuripsw/2023/wang2023neuripsw-model/)

BibTeX

@inproceedings{wang2023neuripsw-model,
  title     = {{Model Evaluation for Geospatial Problems}},
  author    = {Wang, Jing and Hallman, Tyler and Hopkins, Laurel and Kilbride, John Burns and Robinson, W. Douglas and Hutchinson, Rebecca},
  booktitle = {NeurIPS 2023 Workshops: CompSust},
  year      = {2023},
  url       = {https://mlanthology.org/neuripsw/2023/wang2023neuripsw-model/}
}