Non-parametric Lagrangian biasing from the insights of neural nets

Abstract

We present a Lagrangian model of galaxy clustering bias in which we train a neural net using the local properties of the smoothed initial density field to predict the late-time mass-weighted halo field. By fitting the mass-weighted halo field in the AbacusSummit simulations at z=0.5, we find that including three coarsely spaced smoothing scales gives the best recovery of the halo power spectrum. Adding more smoothing scales may lead to 2-5% underestimation of the large-scale power and can cause the neural net to overfit. We find that the fitted halo-to-mass ratio can be well described by two directions in the original high-dimension feature space. Projecting the original features into these two principal components and re-training the neural net either reproduces the original training result, or outperforms it with a better match of the halo power spectrum. The elements of the principal components are unlikely to be assigned physical meanings, partly owing to the features being highly correlated between different smoothing scales. Our work illustrates a potential need to include multiple smoothing scales when studying galaxy bias, and this can be done easily with machine-learning methods that can take in high dimensional input feature space.

arXiv commentsubmitted to JCAP

References in corpus (40)

• Large scale bias and the peak background split
• An analytic model for the spatial clustering of dark matter haloes
• The DESI Experiment Part I: Science,Targeting, and Survey Design
• Large-Scale Galaxy Bias
• Biasing and Hierarchical Statistics in Large-scale Structure
• Clustering of dark matter tracers: generalizing bias for the coming era of precision LSS
• Gravity and Large-Scale Non-local Bias
• Evidence for Quadratic Tidal Tensor Bias from the Halo Bispectrum
• Correcting for the alias effect when measuring the power spectrum using FFT
• AbacusSummit: A Massive Set of High-Accuracy, High-Resolution $N$-Body Simulations
• Renormalized Halo Bias
• Understanding higher-order nonlocal halo bias at large scales by combining the power spectrum with the bispectrum
• Precision measurement of the local bias of dark matter halos
• Modeling scale-dependent bias on the baryonic acoustic scale with the statistics of peaks of Gaussian random fields
• The $\text{Abacus}$ Cosmological $N$-body Code
• Modeling Biased Tracers at the Field Level
• Limitations to the "basic'' HOD model and beyond
• The Abacus Cosmos: A Suite of Cosmological N-body Simulations
• The Gaussian streaming model and Lagrangian effective field theory
• A High-Fidelity Realization of the Euclid Code Comparison $N$-body Simulation with Abacus
• Cubic Halo Bias in Eulerian and Lagrangian Space
• The cosmology dependence of galaxy clustering and lensing from a hybrid $N$-body-perturbation theory model
• \textsc{CompaSO}: A new halo finder for competitive assignment to spherical overdensities
• Simulations and symmetries
• Beyond LIMD bias: a measurement of the complete set of third-order halo bias parameters
• Improving Initial Conditions for Cosmological $N$-Body Simulations
• Modeling Galaxies in Redshift Space at the Field Level
• Halo bias in Lagrangian Space: Estimators and theoretical predictions
• Machine learning cosmological structure formation
• Reconstructing large-scale structure with neutral hydrogen surveys
• Galaxy bias from forward models: linear and second-order bias of IllustrisTNG galaxies
• Extensions to models of the galaxy-halo connection
• Very Massive Tracers and Higher Derivative Biases
• Scale dependent halo bias in the excursion set approach
• Exploring the squeezed three-point galaxy correlation function with generalized halo occupation distribution models
• An interpretable machine learning framework for dark matter halo formation
• Velocity bias in the distribution of dark matter halos
• A robust measurement of the first higher-derivative bias of dark matter halos
• Non-local bias contribution to third-order galaxy correlations
• A fully Lagrangian, non-parametric bias model for dark-matter halos

Cited by in corpus (1)

• Gaussian Lagrangian Galaxy Bias