Point Process Learning: a cross-validation-based statistical framework for point processes

Paper in proceeding, 2025

Recently, Point Process Learning was introduced as a powerful approach to fitting Papangelou conditional intensity models to point pattern data. This cross-validation-based statistical theory was shown to significantly outperform the state-of-the-art in the context of kernel intensity estimation. In this paper, we further illustrate its potential by showing that it outperforms the state-of-the-art when fitting a hard-core Gibbs model.