arXiv subjects

Carolina Cuesta-Lazaro

Publications and source records attributed to Carolina Cuesta-Lazaro.

At least 19 recordsLinked to original sources

arXiv2026

Feedback-Conditional 3D Reconstruction of Cosmic Baryons with Flow Matching

The complex relationship between galaxies and diffuse gas at large scales depends strongly on astrophysical feedback and cosmology. The dispersion measure (DM) of Fast Radio Bursts (FRBs) is a column of cosmic baryon density and is dominated by the intergalactic medium, which is not necessarily captured by traditional halo model prescriptions. We introduce a generative model to reconstruct 3D gas density fields from galaxy surveys, including redshift space distortions. The probabilistic model is trained on the CAMELS50 simulation suite (IllustrisTNG) and learns to predict the posterior distribution of gas fields conditioned on stellar mass-weighted galaxy density, cosmological parameters, the minimum stellar mass threshold for the galaxy sample, and the baryon suppression power spectrum as a measure of baryonic feedback. Our method uses the 3D gas reconstruction to estimate per-sightline FRB DMs and their posteriors. We validate the model on test boxes by showing agreement between predicted and true DM across individual lines of sight and the overall DM distribution. The model generalizes to a box size $\sim$\,4$^3$ times larger in volume than the training boxes and can cover the volumes spanned by cosmological FRB sightlines. The model's 3D gas density predictions vary with input power spectrum and feedback scenarios, offering an inference path to constrain feedback models from FRBs and other probes such as X-rays, the SZ effect, and weak lensing.

astro-ph.CO↗

arXiv2026

From starlight to dark matter: a stochastic interpolation approach to map dark matter from stellar density

The dark matter halo profile in galaxies holds key information about the nature of dark matter and galaxy formation. Constraining the dark matter profile of galaxies beyond the Milky Way traditionally requires expensive spectroscopic observations for kinematic information. In this paper, we explore a conditional generative model framework to map the dark matter profile of Milky Way-mass galaxies from stellar density profiles, more easily obtainable through large photometric imaging surveys. As a proof of concept, we train the model to learn a stochastic bridge between instrument systematics-free baryonic stellar distributions and underlying dark matter density maps from the DREAMS hydrodynamics simulation suite. We recover 2D dark matter density profiles with a typical accuracy of $\sim0.1$ dex ($\sim1.5$% of the truth in log scale). The stochastic sampling procedure provides uncertainty estimates of the predicted dark matter map, with typical values $\sim0.1$ dex. Out-of-domain tests with Milky Way-mass galaxies from IllustrisTNG and FIRE simulations show that, while the model can qualitatively be generalized to TNG50 galaxies from IllustrisTNG, the model is sensitive to the galaxy formation model, with $\sim0.2$-$0.4$ dex over-prediction for the inner profiles ($r\lesssim5$ kpc) of the FIRE test galaxies. Future work will explore training with additional suites of simulations and/or conditioning on additional information, such as multi-band images. Our results are a first step towards using generative models as a flexible, uncertainty-aware framework for turning forthcoming data from large imaging surveys into spatially resolved dark matter maps.

astro-ph.GA↗

arXiv2026

Unknown Unknowns: Model Misspecification in Machine Learning for Physics

Machine learning is now a central tool for solving inverse problems in particle physics and astronomy. Models are trained on simulation and deployed on real data, raising the question not just of whether they fit, but of whether they are wrong in ways we did not anticipate: the unknown unknowns. This challenge of model misspecification is not unique to machine learning. In physics, misspecification is sometimes exactly what we want to find: new discoveries appear as failures of existing models. At other times, we want such effects absorbed into the analysis without biasing the measurement. A robust analysis is one that absorbs the misspecifications we are not interested in, while preserving sensitivity to the ones we are. Machine learning can both amplify misspecification and provide new tools to address it. We discuss the challenges of model misspecification, diagnostics for detecting it, and strategies for mitigation. No single diagnostic can confirm that a model is correctly specified: detection and mitigation are two halves of an iterative loop, in which a battery of complementary diagnostics is applied, the model is updated, and the process repeated. Robustness against unknown unknowns is ultimately less about any single technique than about a disposition: a willingness to suspect one's own model, and to design analyses that can survive being wrong in ways one did not anticipate.

physics.data-an↗

arXiv2026

A Parameter-Masked Mock Data Challenge for Beyond-Two-Point Galaxy Clustering Statistics

The last few years have seen the emergence of a wide array of novel techniques for analyzing high-precision data from upcoming galaxy surveys, which aim to extend the statistical analysis of galaxy clustering data beyond the linear regime and the canonical two-point (2pt) statistics. We test and benchmark some of these new techniques in a community data challenge "Beyond-2pt", initiated during the Aspen 2022 Summer Program "Large-Scale Structure Cosmology beyond 2-Point Statistics," whose first round of results we present here. The challenge dataset consists of high-precision mock galaxy catalogs for clustering in real space, redshift space, and on a light cone. Participants in the challenge have developed end-to-end pipelines to analyze mock catalogs and extract unknown ("masked") cosmological parameters of the underlying $Λ$CDM models with their methods. The methods represented are density-split clustering, nearest neighbor statistics, BACCO power spectrum emulator, void statistics, LEFTfield field-level inference using effective field theory (EFT), and joint power spectrum and bispectrum analyses using both EFT and simulation-based inference. In this work, we review the results of the challenge, focusing on problems solved, lessons learned, and future research needed to perfect the emerging beyond-2pt approaches. The unbiased parameter recovery demonstrated in this challenge by multiple statistics and the associated modeling and inference frameworks supports the credibility of cosmology constraints from these methods. The challenge data set is publicly available and we welcome future submissions from methods that are not yet represented.

astro-ph.CO↗

arXiv2026

Learning What's Real: Disentangling Signal and Measurement Artifacts in Multi-Sensor Data, with Applications to Astrophysics

Data collected from the physical world is always a combination of multiple sources: an underlying signal from the physical process of interest and a signal from measurement-dependent artifacts from the sensor or instrument. This secondary signal acts as a confounding factor, limiting our ability to extract information about the physics underlying the phenomena we observe. Furthermore, it complicates the combination of observations in heterogeneous or multi-instrument settings. We propose a deep learning framework that leverages overlapping observations, a dual-encoder architecture, and a counterfactual generation objective to disentangle these factors of variation. The resulting representations explicitly separate intrinsic signals from sensor-specific distortions and noise, and can be used for counterfactual view generation, parameter inference unconfounded by measurement distortions, and instrument-independent similarity search. We demonstrate the effectiveness of our approach on astrophysical galaxy images from the DESI Legacy Imaging Survey (Legacy) and the Hyper Suprime-Cam (HSC) Survey as a representative multi-instrument setting. This framework provides a general recipe for scientific and multi-modal self-supervised pretraining: construct training pairs from overlapping observations of the same physical system, treat sensor- or modality-specific effects as augmentations, and learn invariant representations through counterfactual generation.

astro-ph.IM↗

arXiv2025

BaryonBridge: Stochastic Interpolant Model for Fast Hydrodynamical Simulations

Constructing a general-purpose framework for mapping between dark matter simulations and observable hydrodynamical simulation outputs is a long-standing problem in modern astrophysics. In this work, we present a new approach utilizing stochastic interpolants to map between cheap fast particle mesh simulations and baryonic quantities in three dimensions, requiring a total of 7 GPU minutes per 256^3 grid size simulation. Using the CAMELS multifield dataset, we are able to condition our mapping on both cosmological and astrophysical properties. We focus this work on hydrodynamical quantities suitable for Lya observables finding excellent agreement up to small spatial scales, k ~ 10.0 (h^(-1) Mpc) at z=2.0, for Lya flux statistics. Our approach is fully convolutional, allowing training on comparatively small volumes and application to larger volumes, which was tested on TNG50.

astro-ph.CO↗

arXiv2025

Hierarchical Simulation-Based Inference of Supernova Power Sources and their Physical Properties

Time domain surveys such as the Vera C. Rubin Observatory are projected to annually discover millions of astronomical transients. This and complementary programs demand fast, automated methods to constrain the physical properties of the most interesting objects for spectroscopic follow up. Traditional approaches to likelihood-based inference are computationally expensive and ignore the multi-component energy sources powering astrophysical phenomena. In this work, we present a hierarchical simulation-based inference model for multi-band light curves that 1) identifies the energy sources powering an event of interest, 2) infers the physical properties of each subclass, and 3) separates physical anomalies in the learned embedding space. Our architecture consists of a transformer-based light curve summarizer coupled to a flow-matching regression module and a categorical classifier for the physical components. We train and test our model on $\sim$150k synthetic light curves generated with $\texttt{MOSFiT}$. Our network achieves a 90% classification accuracy at identifying energy sources, yields well-calibrated posteriors for all active components, and detects rare anomalies such as tidal disruption events (TDEs) through the learned latent space. This work demonstrates a scalable joint framework for population studies of known transients and the discovery of novel populations in the era of Rubin.

astro-ph.IM↗

arXiv2025

Detecting Model Misspecification in Cosmology with Scale-Dependent Normalizing Flows

Current and upcoming cosmological surveys will produce unprecedented amounts of high-dimensional data, which require complex high-fidelity forward simulations to accurately model both physical processes and systematic effects which describe the data generation process. However, validating whether our theoretical models accurately describe the observed datasets remains a fundamental challenge. An additional complexity to this task comes from choosing appropriate representations of the data which retain all the relevant cosmological information, while reducing the dimensionality of the original dataset. In this work we present a novel framework combining scale-dependent neural summary statistics with normalizing flows to detect model misspecification in cosmological simulations through Bayesian evidence estimation. By conditioning our neural network models for data compression and evidence estimation on the smoothing scale, we systematically identify where theoretical models break down in a data-driven manner. We demonstrate a first application to our approach using matter and gas density fields from three CAMELS simulation suites with different subgrid physics implementations.

astro-ph.CO↗

arXiv2025

CosmoFlow: Scale-Aware Representation Learning for Cosmology with Flow Matching

Generative machine learning models have been demonstrated to be able to learn low dimensional representations of data that preserve information required for downstream tasks. In this work, we demonstrate that flow matching based generative models can learn compact, semantically rich latent representations of field level cold dark matter (CDM) simulation data without supervision. Our model, CosmoFlow, learns representations 32x smaller than the raw field data, usable for field level reconstruction, synthetic data generation, and parameter inference. Our model also learns interpretable representations, in which different latent channels correspond to features at different cosmological scales.

astro-ph.CO↗

arXiv2025

A theoretical approach to density-split clustering

We present an analytical model for density-split correlation functions, that probe galaxy clustering in different density environments. Specifically, we focus on the cross-correlation between density-split regions and the tracer density field. We show that these correlation functions can be expressed in terms of the two-point probability density function (PDF) of the density field. We derive analytical predictions using three levels of approximation for the two-point PDF: a bivariate Gaussian distribution, a bivariate shifted log-normal distribution, and a prediction based on the Large Deviation Theory (LDT) framework. For count-in-cell densities, obtained through spherical top-hat smoothing, one can leverage spherical collapse dynamics and LDT to predict the density two-point PDF in the large-separation regime relative to the smoothing radius. We validate our model against dark matter N-body simulations in real space, incorporating Poisson shot noise and galaxy bias. Our results show that the LDT prediction outperforms the log-normal approximation, and agrees with simulations on large scales within the cosmic variance of a typical DESI DR1 sample, despite relying on only one degree of freedom.

astro-ph.CO↗

arXiv2025

Cosmological constraints from the Minkowski functionals of the BOSS CMASS galaxy sample

For the first time, we develop a simulation-based model for the Minkowski functionals (MFs) of large-scale structure, which allows us to extract the full information available from the MFs (including both the Gaussian and non-Gaussian part), and apply it to the BOSS DR12 CMASS galaxy sample. Our model is based on high-fidelity mock galaxy catalogs constructed from the \textsc{Abacus}\textsc{Summit} simulations using the halo occupation distribution (HOD) framework, which include the redshift-space distortions and Alcock-Paczynski distortions, incorporate survey realism, including survey geometry and veto masks, and account for angular plus radial selection effects. The cosmological and HOD parameter dependence of the MFs is captured with a neural network emulator trained from the galaxy mocks with various cosmological and HOD parameters. To benchmark the constraining power of the MFs, we also train an emulator for the galaxy 2-point correlation function (2PCF) using the same pipeline. Having validated our approach through successful parameter recovery tests on both internal and external mocks, including non-HOD forward models of the halo-galaxy connection, we apply our forward model to analyze the CMASS data in the redshift range $0.45<z<0.58$. We find the MFs provide stronger constraints on the cosmological parameters than the 2PCF. The combination of the two gives $ω_{\rm cdm}=0.1172^{+0.0020}_{-0.0023}$, $σ_8=0.783\pm 0.026$, and $n_s=0.966^{+0.019}_{-0.015}$, which are tighter by a factor of 2.0, 1.9, and 1.6 than the 2PCF alone. The derived constraint $fσ_8=0.453 \pm 0.016$ is also improved by a factor of 1.9, compared to the 2PCF, and agrees well with Planck 2018 predictions and other results from a series of studies in the literature.

astro-ph.CO↗

arXiv2025

High-redshift Millennium and Astrid galaxies in effective field theory at the field level

Effective Field Theory (EFT) modeling is expected to be a useful tool in the era of future higher-redshift galaxy surveys such as DESI-II and Spec-S5 due to its robust description of various large-scale structure tracers. However, large values of EFT bias parameters of higher-redshift galaxies could jeopardize the convergence of the perturbative expansion. In this paper we measure the bias parameters and other EFT coefficients from samples of two types of star-forming galaxies in the state-of-the-art MilleniumTNG and Astrid hydrodynamical simulations. Our measurements are based on the field-level EFT forward model that allows for precision EFT parameter measurements by virtue of cosmic variance cancellation. Specifically, we consider approximately representative samples of Lyman-break galaxies (LBGs) and Lyman-alpha emitters (LAEs) that are consistent with the observed (angular) clustering and number density of these galaxies at $z=3$. Reproducing the linear biases and number densities observed from existing LAE and LBG data, we find quadratic bias parameters that are roughly consistent with those predicted from the halo model coupled with a simple halo occupation distribution model. We also find non-perturbative velocity contributions (Fingers of God) of a similar size for LBGs to the familiar case of Luminous Red Galaxies. However, these contributions are quite small for LAEs despite their large satellite fraction values of up to $\sim 30\%$. Our results indicate that the effective momentum reach $k_{\rm{Max}}$ at $z=3$ for LAEs (LBGs) will be in the range $0.3-0.6 ~h\rm{Mpc}^{-1}$ ($0.2-0.8~h\rm{Mpc}^{-1}$), suggesting that EFT will perform well for high redshift galaxy clustering. This work provides the first step toward obtaining realistic simulation-based priors on EFT parameters for LAEs and LBGs.

astro-ph.CO↗

arXiv2025

Full-shape analysis with simulation-based priors: cosmological parameters and the structure growth anomaly

We explore full-shape analysis with simulation-based priors, which is the simplest approach to galaxy clustering data analysis that combines effective field theory (EFT) on large scales and numerical simulations on small scales. The core ingredient of our approach is the prior density of EFT parameters which we extract from a suite of 10500 galaxy simulations based on the halo occupation distribution (HOD) model. We measure the EFT parameters with the field-level forward model, which enables us to cancel cosmic variance. On the theory side, we develop a new efficient approach to calculate field-level transfer functions using time-sliced perturbation theory and the logarithmic fast Fourier transform. We find that the cosmology dependence of EFT parameters of galaxies is approximately degenerate with the HOD parameters, and hence it can be ignored for the purpose of prior generation. We use neural density estimation to model the measured distribution of EFT parameters. Our distribution model is then used as a prior in a reanalysis of the BOSS full-shape galaxy power spectrum data. Assuming the $Λ$CDM model, we find significant ($\approx 30\%$ and $\approx 60\%$) improvements for the matter density fraction and the mass fluctuation amplitude, which are constrained to $Ω_{m}= 0.315 \pm 0.010$ and $σ_8 = 0.671 \pm 0.027$. The value of the Hubble constant does not change, $H_0= 68.7\pm 1.1$~km/s/Mpc. This reaffirms earlier reports of the structure growth tension from the BOSS data. Finally, we use the measured EFT parameters to constrain the galaxy-dark matter connection.

astro-ph.CO↗

arXiv2025

Diffusion-HMC: Parameter Inference with Diffusion-model-driven Hamiltonian Monte Carlo

Diffusion generative models have excelled at diverse image generation and reconstruction tasks across fields. A less explored avenue is their application to discriminative tasks involving regression or classification problems. The cornerstone of modern cosmology is the ability to generate predictions for observed astrophysical fields from theory and constrain physical models from observations using these predictions. This work uses a single diffusion generative model to address these interlinked objectives -- as a surrogate model or emulator for cold dark matter density fields conditional on input cosmological parameters, and as a parameter inference model that solves the inverse problem of constraining the cosmological parameters of an input field. The model is able to emulate fields with summary statistics consistent with those of the simulated target distribution. We then leverage the approximate likelihood of the diffusion generative model to derive tight constraints on cosmology by using the Hamiltonian Monte Carlo method to sample the posterior on cosmological parameters for a given test image. Finally, we demonstrate that this parameter inference approach is more robust to small perturbations of noise to the field than baseline parameter inference networks.

astro-ph.CO↗

arXiv2024

A point cloud approach to generative modeling for galaxy surveys at the field level

We introduce a diffusion-based generative model to describe the distribution of galaxies in our Universe directly as a collection of points in 3-D space (coordinates) optionally with associated attributes (e.g., velocities and masses), without resorting to binning or voxelization. The custom diffusion model can be used both for emulation, reproducing essential summary statistics of the galaxy distribution, as well as inference, by computing the conditional likelihood of a galaxy field. We demonstrate a first application to massive dark matter haloes in the Quijote simulation suite. This approach can be extended to enable a comprehensive analysis of cosmological data, circumventing limitations inherent to summary statistic -- as well as neural simulation-based inference methods.

astro-ph.CO↗

arXiv2024

Emulation of $f(R)$ modified gravity from $Λ$CDM using conditional GANs

A major aim of cosmological surveys is to test deviations from the standard $Λ$CDM model, but the full scientific value of these surveys will only be realised through efficient simulation methods that keep up with the increasing volume and precision of observational data. $N$-body simulations of modified gravity (MG) theories are computationally expensive since highly non-linear equations must be solved. This represents a significant bottleneck in the path to reach the data volume and resolution attained by equivalent $Λ$CDM simulations. We develop a field-level neural network-based emulator that generates density and velocity divergence fields under the $f(R)$ gravity MG model from the corresponding $Λ$CDM simulated fields. Using attention mechanisms and a complementary frequency-based loss function, our model is able to learn this intricate mapping. We use the idea of latent space extrapolation to generalise our emulator to $f(R)$ models with differing field strengths. The predictions of our emulator agree with the $f(R)$ simulations to within 5% for matter density and to within 10% for velocity divergence power spectra up to $k \sim 2\,h$ $\mathrm{Mpc}^{-1}$. But for a few select cases, higher-order statistics are reproduced with $\lesssim$10% agreement. Latent extrapolation allows our emulator to generalise to different parameterisations of the $f(R)$ model without explicitly training on those variants. Given a $Λ$CDM simulation, the GPU-based emulator can reproduce the equivalent $f(R)$ realisation $\sim$600 times faster than full $N$-body simulations. This lays the foundations for a valuable tool for realistic yet rapid mock field generation and robust cosmological analyses.

astro-ph.CO↗

arXiv2024

The Millennium and Astrid galaxies in effective field theory: comparison with galaxy-halo connection models at the field level

Cosmological analyses of redshift space clustering data are primarily based on using luminous ``red'' galaxies (LRGs) and ``blue'' emission line galaxies (ELGs) to trace underlying dark matter. Using the large high-fidelity high-resolution MillenniumTNG (MTNG) and Astrid simulations, we study these galaxies with the effective field theory (EFT)-based field level forward model. We confirm that both red and blue galaxies can be accurately modeled with EFT at the field level and their parameters match those of the phenomenological halo-based models. Specifically, we consider the state of the art Halo Occupation Distribution (HOD) and High Mass Quenched (HMQ) models for the red and blue galaxies, respectively. Our results explicitly confirm the validity of the halo-based models on large scales beyond the two-point statistics. In addition, we validate the field-level HOD/HMQ-based priors for EFT full-shape analysis. We find that the local bias parameters of the ELGs are in tension with the predictions of the LRG-like HOD models and present a simple analytic argument explaining this phenomenology. We also confirm that ELGs exhibit weaker non-linear redshift-space distortions (``fingers-of-God''), suggesting that a significant fraction of their data should be perturbative. We find that the response of EFT parameters to galaxy selection is sensitive to assumptions about baryonic feedback, suggesting that a detailed understanding of feedback processes is necessary for robust predictions of EFT parameters. Finally, using neural density estimation based on paired HOD-EFT parameter samples, we obtain optimal HOD models that reproduce the clustering of Astrid and MTNG galaxies.

astro-ph.CO↗

arXiv2024

A Cosmic-Scale Benchmark for Symmetry-Preserving Data Processing

Efficiently processing structured point cloud data while preserving multiscale information is a key challenge across domains, from graphics to atomistic modeling. Using a curated dataset of simulated galaxy positions and properties, represented as point clouds, we benchmark the ability of graph neural networks to simultaneously capture local clustering environments and long-range correlations. Given the homogeneous and isotropic nature of the Universe, the data exhibits a high degree of symmetry. We therefore focus on evaluating the performance of Euclidean symmetry-preserving ($E(3)$-equivariant) graph neural networks, showing that they can outperform non-equivariant counterparts and domain-specific information extraction techniques in downstream performance as well as simulation-efficiency. However, we find that current architectures fail to capture information from long-range correlations as effectively as domain-specific baselines, motivating future work on architectures better suited for extracting long-range information.

cs.LG↗