Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Particle gradient descent model for point process generation

Antoine Brochard 1, 2 Bartłomiej Błaszczyszyn 1 Stéphane Mallat 3, 4, 5 Sixin Zhang 6
1 DYOGENE - Dynamics of Geometric Networks
DI-ENS - Département d'informatique de l'École normale supérieure, CNRS - Centre National de la Recherche Scientifique : UMR 8548, Inria de Paris
6 IRIT-SC - Signal et Communications
IRIT - Institut de recherche en informatique de Toulouse
Abstract : This paper introduces a generative model for planar point processes in a square window, built upon a single realization of a stationary, ergodic point process observed in this window. Inspired by recent advances in gradient descent methods for maximum entropy models, we propose a method to generate similar point patterns by jointly moving particles of an initial Poisson configuration towards a target counting measure. The target measure is generated via a deterministic gradient descent algorithm, so as to match a set of statistics of the given, observed realization. Our statistics are estimators of the multi-scale wavelet phase harmonic covariance, recently proposed in image modeling. They allow one to capture geometric structures through multi-scale interactions between wavelet coefficients. Both our statistics and the gradient descent algorithm scale better with the number of observed points than the classical k-nearest neighbour distances previously used in generative models for point processes, based on the rejection sampling or simulated-annealing. The overall quality of our model is evaluated on point processes with various geometric structures through spectral and topological data analysis.
Complete list of metadata

Cited literature [36 references]  Display  Hide  Download
Contributor : Antoine Brochard <>
Submitted on : Tuesday, October 27, 2020 - 3:00:53 PM
Last modification on : Tuesday, May 4, 2021 - 4:07:52 PM
Long-term archiving on: : Thursday, January 28, 2021 - 7:07:35 PM


Files produced by the author(s)


  • HAL Id : hal-02980486, version 1


Antoine Brochard, Bartłomiej Błaszczyszyn, Stéphane Mallat, Sixin Zhang. Particle gradient descent model for point process generation. 2020. ⟨hal-02980486⟩



Record views


Files downloads