21cmEMUv3: a hybrid diffusion-LSTM emulator of 21cmFAST summary observables

• URL: http://arxiv.org/abs/2606.00219v1
• Date: Fri, 29 May 2026 18:00:05 GMT
• Title: 21cmEMUv3: a hybrid diffusion-LSTM emulator of 21cmFAST summary observables
• Abstract summary: We present 21cmEMUv3, an emulator trained on 21cmFASTv3 simulations that model both atomically and molecularly cooling galaxies.<n>21cmEMUv3 is conditioned on $_8$ and ten astrophysical parameters to produce seven summary observables.<n>We use the emulator to reinterpret current 21cm PS upper limits from HERA, for the first time using state-of-the-art hydrodynamical simulations.
• Score: 1.631115063641726
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We are witnessing a surge in observations of the cosmic dawn (CD) and epoch of reionisation (EoR), driving an increasing demand for fast and robust theoretical interpretation frameworks. In response, machine learning (ML), and emulation in particular, has emerged as a powerful approach to accelerate and enhance inference pipelines. In this work, we present 21cmEMUv3, an emulator trained on 21cmFASTv3 simulations that model both atomically and molecularly cooling galaxies. 21cmEMUv3 is conditioned on $σ_8$ and ten astrophysical parameters to produce seven summary observables: (i) the cylindrical 21cm power spectrum (PS), emulated for the first time at such high resolution and accuracy across a wide redshift range of $z \sim$ 6--30; (ii) the spherically-averaged 21cm PS; (iii) the mean neutral fraction of the intergalactic medium (IGM); (iv) the mean 21cm spin temperature; (v) the global 21cm signal; (vi) the ultraviolet (UV) luminosity functions (LFs); and (vii) the Thomson scattering optical depth. Notably, the cylindrical 21cm PS is emulated via score-based diffusion, while the remaining six summaries are emulated via long-short term memory (LSTM) networks, all achieving sub-percent median accuracy. We use the emulator to reinterpret current 21cm PS upper limits from HERA, for the first time using state-of-the-art hydrodynamical simulations to inform priors on star formation inside molecularly cooling galaxies. We find that our inferred soft-band X-ray luminosity per unit star formation rate is consistent with extrapolations of high-mass X-ray binaries to the low-metallicity regimes expected in the first galaxies, excluding values below $10^{39.2}$ erg s$^{-1}M^{-1}_\odot \rm{yr}$ at $95\%$ confidence. Finally, we produce forecasts for the detection of the cosmic 21cm PS with the Square Kilometre Array for different array configurations. The 21cmEMU package is publicly available.

Related papers

• Ultrafast and high resolution spatial light modulation for cold atoms [70.8440518934986]
  We develop and numerically validate a scheme for fully programmable Hubbard models, with time-dependent control over local chemical potentials, tunneling amplitudes, on-site interactions, and patterns of artificial magnetic flux.<n>The same architecture performs fast, arbitrary permutations of tweezers in 2D, decoupling optical constraints from the design of high-rate error-correcting codes.
  arXiv  Detail & Related papers  (2026-08-18T17:56:25Z)
• Precision quantum simulation of magnon spectra and interactions [94.30337065819249]
  We show high-precision simulation of both linear and non-linear response functions in a 2D XY spin-1/2 magnet using an analog-digital superconducting processor of up to 97 qubits.<n>Our results reveal stark variations in magnon decay rates across the Brillouin zone, with enhancement near van Hove singularities and suppression for edge-localized modes.
  arXiv  Detail & Related papers  (2026-07-14T22:11:22Z)
• Three-dimensional Conditional Diffusion Models for Cosmological 21 cm Lightcone Emulation [4.210446607355696]
  We investigate conditional diffusion modeling for three-dimensional 21 cm lightcone emulation.<n>We focus on cubes with a sky-plane size of $64times64$ and a line-of-sight depth up to 1024 cells.<n>We perform controlled comparisons across preprocessing choices, dynamic-range compression settings, architecture depth, and training duration.
  arXiv  Detail & Related papers  (2026-05-27T19:16:49Z)
• The Spectral Lifecycle of Transformer Training: Transient Compression Waves, Persistent Spectral Gradients, and the Q/K--V Asymmetry [4.28787537081191]
  We present the first systematic study of weight matrix singular value spectra emphduring transformer pretraining.<n>We discover three phenomena: textbf(1)Transient Compression Waves: stable rank compression propagates as a traveling wave from early to late layers.<n>textbf(2)Persistent Spectral Gradients: the power-law exponent$$ develops a permanent depth gradient forming a non-monotonic inverted-U in deeper models.
  arXiv  Detail & Related papers  (2026-04-03T08:58:53Z)
• Cosmo-FOLD: Fast generation and upscaling of field-level cosmological maps with overlap latent diffusion [1.9116784879310027]
  We introduce a novel technique to rapidly generate accurate and arbitrarily large cosmological and astrophysical 3-dimensional fields.<n>We are able to generate TNG300-2 dark matter density and gas temperature fields from a model trained only on 1% of the volume.<n>These results are obtained within a small fraction of the original simulation cost and produced on a single GPU.
  arXiv  Detail & Related papers  (2026-01-20T19:00:04Z)
• Quantum Ising Model on $(2+1)-$Dimensional Anti$-$de Sitter Space using Tensor Networks [37.108493798440655]
  We study the quantum Ising model on (2+1)-dimensional anti-de Sitter space using Matrix Product States (MPS) and Matrix Product Operators (MPOs)<n>Our spatial lattices correspond to regular tessellations of hyperbolic space with coordination number seven.<n>We find the ground state of this model using the Density Matrix Renormalization Group (DMRG) algorithm which allowed us to probe lattices that range in size up to 232 sites.
  arXiv  Detail & Related papers  (2025-12-23T23:29:39Z)
• Quantum Simulation of Electron Energy Loss Spectroscopy for Battery Materials [38.065424255210765]
  We present a quantum algorithm and an end-to-end simulation framework to compute the dynamic structure factor (DSF)<n>We apply this approach to the simulation of electron energy loss spectroscopy (EELS) in the core-level electronic excitation regime.
  arXiv  Detail & Related papers  (2025-08-21T20:01:16Z)
• Accelerating exoplanet climate modelling: A machine learning approach to complement 3D GCM grid simulations [39.58317527488534]
  This study aims to determine whether machine learning (ML) algorithms can be used to predict the 3D temperature and wind structure of exoplanets.<n>A new 3D GCM grid with 60 inflated hot Jupiters orbiting A, F, G, K, and M-type host stars modelled with Exorad has been introduced.<n>A dense neural network (DNN) and a decision tree algorithm (XGBoost) are trained on this grid to predict local gas temperatures along with horizontal and vertical winds.
  arXiv  Detail & Related papers  (2025-08-14T16:50:38Z)
• Predictive Hydrodynamic Simulations for Laser Direct-drive Implosion Experiments via Artificial Intelligence [45.84035159987504]
  This work presents predictive hydrodynamic simulations empowered by artificial intelligence (AI) for laser driven implosion experiments.<n>A Transformer-based deep learning model MULTI-Net is established to predict implosion features according to laser waveforms and target radius.<n>This study demonstrates a data-driven AI framework that enhances the prediction ability of simulations for complicated laser fusion experiments.
  arXiv  Detail & Related papers  (2025-07-22T04:57:40Z)
• Exponential onset of scalable entanglement via twist-and-turn dynamics in XY models [41.94295877935867]
  We show that the so-called twist-and-turn" (TaT) dynamics can offer an important resource to reach scalable multipartite entanglement.<n>For dipolar interactions, the entanglement dynamics at intermediate times is completely at odds with thermalization.
  arXiv  Detail & Related papers  (2025-07-10T22:34:44Z)
• Variational diffusion transformers for conditional sampling of supernovae spectra [49.1574468325115]
  We introduce DiTSNe-Ia, a variational diffusion-based generative model conditioned on light curve observations and trained to reproduce the observed spectral diversity of SNe Ia.<n>In experiments with realistic light curves and spectra from radiative transfer simulations, DiTSNe-Ia achieves significantly more accurate reconstructions than the widely used SALT3 templates.
  arXiv  Detail & Related papers  (2025-05-05T22:52:32Z)
• 21cmEMU: an emulator of 21cmFAST summary observables [0.0]
  We present 21cmEMU: an emulator of several summary observables from the popular 21cmFAST simulation code. 21cmEMU takes as input nine parameters characterizing EoR galaxies, and outputs the following summary statistics. All observables are predicted with sub-percent median accuracy, with a reduction of the computational cost by a factor of over 10$4$.
  arXiv  Detail & Related papers  (2023-09-11T18:00:00Z)
• A unified framework for 21cm tomography sample generation and parameter inference with Progressively Growing GANs [7.6146285961466]
  We utilise a specific type of neural network, a Progressively Growing Generative Adversarial Network (PGGAN), to produce realistic tomography images of the 21cm brightness temperature during the Epoch of Reionisation (EoR) The network generates new samples at a resolution of $sim 3'$ in a second (on a laptop CPU), and the resulting global 21cm signal, power spectrum, and pixel distribution function agree well with those of the training data.
  arXiv  Detail & Related papers  (2020-02-19T00:39:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.