Reconstructing the local density field with combined convolutional and point cloud architecture

Baptiste Barthe-Gold, Nhat-Minh Nguyen, Leander Thiele
10/9/2025
astro-ph.COcs.LGstat.ML

No confident code implementation found for this paper.

Abstract

We construct a neural network to perform regression on the local dark-matter density field given line-of-sight peculiar velocities of dark-matter halos, biased tracers of the dark matter field. Our architecture combines a convolutional U-Net with a point-cloud DeepSets. This combination enables efficient use of small-scale information and improves reconstruction quality relative to a U-Net-only approach. Specifically, our hybrid network recovers both clustering amplitudes and phases better than the U-Net on small scales.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Explaining dark matter halo density profiles with neural networks

83% match2023

Fast emulation of cosmological density fields based on dimensionality reduction and supervised machine-learning

79% match2023

Estimating Dark Matter Halo Masses in Simulated Galaxy Clusters with Graph Neural Networks

78% match2024

A deep-learning algorithm to disentangle self-interacting dark matter and AGN feedback models

78% match2024

Hybrid summary statistics: neural weak lensing inference beyond the power spectrum

78% match1 repo22024

Learning the Universe: Learning to Optimize Cosmic Initial Conditions with Non-Differentiable Structure Formation Models

78% match2025

Cite this paper

@article{barthegold2025reconstructing,
  title  = {Reconstructing the local density field with combined convolutional and point cloud architecture},
  author = {Baptiste Barthe-Gold and Nhat-Minh Nguyen and Leander Thiele},
  year   = {2025},
  eprint = {2510.08573},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2510.08573v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2510.08573.svg)](https://codeofpaper.com/paper/2510.08573)

Discussion