jFoF: GPU Cluster Finding with Gradient Propagation

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Horowitz, Benjamin, Bayer, Adrian E.
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866908696317001728
author Horowitz, Benjamin
Bayer, Adrian E.
author_facet Horowitz, Benjamin
Bayer, Adrian E.
contents We present jFoF, a fully GPU-native Friends-of-Friends (FoF) halo finder designed for both high-performance simulation analysis and differentiable modeling. Implemented in JAX, jFoF achieves end-to-end acceleration by performing all neighbor searches, label propagation, and group construction directly on GPUs, eliminating costly host--device transfers. We introduce two complementary neighbor-search strategies, a standard k-d tree and a novel linked-cell grid, and demonstrate that jFoF attains up to an order-of-magnitude speedup compared to optimized CPU implementations while maintaining consistent halo catalogs. Beyond performance, jFoF enables gradient propagation through discrete halo-finding operations via both frozen-assignment and topological optimization modes. Using a topological optimization approach via a REINFORCE-style estimator, our approach allows smooth optimization of halo connectivity and membership, bridging continuous simulation fields with discrete structure catalogs. These capabilities make jFoF a foundation for differentiable inference, enabling end-to-end, gradient-based optimization of structure formation models within GPU-accelerated astrophysical pipelines. We make our code publicly available at https://github.com/bhorowitz/jFOF/.
format Preprint
id arxiv_https___arxiv_org_abs_2510_26851
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle jFoF: GPU Cluster Finding with Gradient Propagation
Horowitz, Benjamin
Bayer, Adrian E.
Instrumentation and Methods for Astrophysics
Cosmology and Nongalactic Astrophysics
Astrophysics of Galaxies
We present jFoF, a fully GPU-native Friends-of-Friends (FoF) halo finder designed for both high-performance simulation analysis and differentiable modeling. Implemented in JAX, jFoF achieves end-to-end acceleration by performing all neighbor searches, label propagation, and group construction directly on GPUs, eliminating costly host--device transfers. We introduce two complementary neighbor-search strategies, a standard k-d tree and a novel linked-cell grid, and demonstrate that jFoF attains up to an order-of-magnitude speedup compared to optimized CPU implementations while maintaining consistent halo catalogs. Beyond performance, jFoF enables gradient propagation through discrete halo-finding operations via both frozen-assignment and topological optimization modes. Using a topological optimization approach via a REINFORCE-style estimator, our approach allows smooth optimization of halo connectivity and membership, bridging continuous simulation fields with discrete structure catalogs. These capabilities make jFoF a foundation for differentiable inference, enabling end-to-end, gradient-based optimization of structure formation models within GPU-accelerated astrophysical pipelines. We make our code publicly available at https://github.com/bhorowitz/jFOF/.
title jFoF: GPU Cluster Finding with Gradient Propagation
topic Instrumentation and Methods for Astrophysics
Cosmology and Nongalactic Astrophysics
Astrophysics of Galaxies
url https://arxiv.org/abs/2510.26851