Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2305.03077

Explaining dark matter halo density profiles with neural networks

Luisa Lucie-Smith·Hiranya V. Peiris·Andrew Pontzen
View original source ↗Read source document ↗Export metadata ↓

Abstract

We use explainable neural networks to connect the evolutionary history of dark matter halos with their density profiles. The network captures independent factors of variation in the density profiles within a low-dimensional representation, which we physically interpret using mutual information. Without any prior knowledge of the halos' evolution, the network recovers the known relation between the early time assembly and the inner profile, and discovers that the profile beyond the virial radius is described by a single parameter capturing the most recent mass accretion rate. The results illustrate the potential for machine-assisted scientific discovery in complicated astrophysical datasets.

Explore related subjects

astro-ph.COcs.LG

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Luisa Lucie-Smith, Hiranya V. Peiris, Andrew Pontzen. 2024-01-19. Explaining dark matter halo density profiles with neural networks. https://doi.org/10.1103/physrevlett.132.031001

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Constraints on the Thomson optical depth to the CMB from the Lyman-$α$ forest

We present the first constraints on the electron optical depth to reionization, $τ_{\mathrm{e}}$, from the Lyman-$α$ forest alone for physically motivated reionization models that match the reionization's end-point, $z_{\rm{end}}$, required by the same astrophysical probe, and for symmetric reionization models with fixed duration, $Δz$, commonly adopted in CMB reionization analyses. Compared to traditional estimates from the latter, the Lyman-$α$ forest traces the ionization state of the IGM through its coupling with the thermal state. We find an explicit mapping between the two solving the chemistry and temperature evolution equations for hydrogen and helium. Our results yield $τ_{\mathrm{e}}$=$0.040^{+0.042}_{-0.018}$ (95\% C.L) and $τ_{\mathrm{e}}$=$0.041^{+0.028}_{-0.017}$ for reionization models with $z_{\rm{end}}$ and $Δz$-fixed, respectively. With mock Lyman-$α$ forest data that mimics the precision of future larger quasar sample datasets, we would potentially obtain tighter $τ_{\mathrm{e}}$ constraints, paving the way for CMB-independent constraints on the epoch of reionization from a large-scale structure probe.

astro-ph.CO↗
arXiv2026

Non-minimally Coupled Running Curvaton for DESI-motivated Dynamical Dark Energy

Recent DESI BAO data combined with CMB and supernova measurements suggest a dynamical dark energy that can cross the phantom divide. We show that introducing a non-minimal coupling $ξχ^2R$ to the running-curvaton framework allows a single field to drive early-universe curvature perturbations and late-time phantom-crossing cosmic acceleration without ghost instabilities. Using MCMC background likelihoods with DESI DR2 BAO, Pantheon+ SNe, and a reduced CMB prior, we constrain the cosmological parameters: $H_0 = 67.88^{+0.53}_{-0.61}\,{\rm km\,s^{-1}\,Mpc^{-1}}$, $Ω_m = 0.3072^{+0.0059}_{-0.0054}$, $w_0 = -0.922^{+0.055}_{-0.063}$, and $w_a = -0.205^{+0.173}_{-0.182}$. Parameter degeneracies leave the coupling constants weakly constrained, highlighting the need for full perturbation-level analysis using CMB spectra and lensing.

astro-ph.CO↗
arXiv2026

On the Relation Between Field-Level Posteriors, Correlators, and their Likelihoods

We develop a field-level posterior for cosmological data by marginalizing over initial conditions and noise in a general forward model. While our focus is on large-scale structure data, the results generalize to any weakly non-Gaussian observable. Moreover, the construction is non-perturbative with respect to the forward model and applies equally well to perturbative calculations, simulation-based predictions, and more general effective descriptions. Expanding the FLP around its Gaussian limit, we derive a general expression for the Fisher matrix and reorganize the field-level information into contributions associated with the connected correlators of the evolved field. This makes explicit which terms are captured by likelihood analyses based on the power spectrum, the bispectrum, or finite sets of summary statistics, and which are lost under compression. We recover the standard Gaussian-covariance result for the power spectrum, show that the Gaussian bispectrum likelihood reproduces the corresponding field-level contribution, and show how cross-covariances among summaries progressively reconstruct more of the full field-level information. As an application to the BAO scale, we show how the field contains all the information required for its optimal reconstruction in the presence of noise, and identify the contributions in the FLP needed to attain this limit. We also show that the reconstruction of the initial field arises naturally as a byproduct of our approach, yielding the optimal estimate of the initial conditions given the data and the noise. Our results provide a unified framework to compare field-level and correlator-based inference, to quantify the information loss induced by compression, and to explore the role of stochasticity.

astro-ph.CO↗