arXiv subjects

Roberto Trotta

Publications and source records attributed to Roberto Trotta.

At least 37 records · Page 2Linked to original sources

arXiv2025

JERALD: high-fidelity dark matter, stellar mass and neutral hydrogen maps from fast N-body simulations

We present a new code and approach, JERALD -- JAX Enhanced Resolution Approximate Lagrangian Dynamics -- , that improves on and extends the Lagrangian Deep Learning method of Dai & Seljak (2021), producing high-resolution dark matter, stellar mass and neutral hydrogen maps from lower-resolution approximate $N$-body simulations. The model is trained using the Sherwood-Relics simulation suite (for a fixed cosmology), specifically designed for the intergalactic medium and the neutral hydrogen distribution in the cosmic web. The output is tested in the redshift range from $z=5$ to $z=0$ and the generalization properties of the learned mapping is demonstrated. JERALD produces maps with dark matter, stellar and neutral hydrogen power spectra in excellent agreement with full-hydrodynamic simulations with $8\times$ higher resolution, at large and intermediate scales; in particular, JERALD's neutral hydrogen power spectra agree with their higher-resolution full-hydrodynamic counterparts within 90% up to $k\simeq1\,h$Mpc$^{-1}$ and within 70% up to $k\simeq10\,h$Mpc$^{-1}$. JERALD provides a fast, accurate and physically motivated approach that we plan to embed in a statistical inference pipeline, such as Simulation-Based Inference, to constrain dark matter properties from large- to intermediate-scale structure observables.

astro-ph.CO↗

arXiv2024

Bayesian evidence estimation from posterior samples with normalizing flows

We propose a novel method ($floZ$), based on normalizing flows, to estimate the Bayesian evidence (and its numerical uncertainty) from a pre-existing set of samples drawn from the unnormalized posterior distribution. We validate it on distributions whose evidence is known analytically, up to 15 parameter space dimensions, and compare with two state-of-the-art techniques for estimating the evidence: nested sampling (which computes the evidence as its main target) and a $k$-nearest-neighbors technique that produces evidence estimates from posterior samples. Provided representative samples from the target posterior are available, our method is more robust to posterior distributions with sharp features, especially in higher dimensions. For a simple multivariate Gaussian, we demonstrate its accuracy for up to 200 dimensions with $10^5$ posterior samples. $floZ$ has wide applicability, e.g., to estimate evidence from variational inference, Markov Chain Monte Carlo samples, or any other method that delivers samples and their likelihood from the unnormalized posterior density. As a physical application, we use $floZ$ to compute the Bayes factor for the presence of the first overtone in the ringdown signal of the gravitational wave data of GW150914, finding good agreement with nested sampling.

stat.ML↗

arXiv2024

The EDGES measurement disfavors an excess radio background during the cosmic dawn

In 2018 the EDGES experiment claimed the first detection of the global cosmic 21cm signal, which featured an absorption trough centered around $z \sim 17$ with a depth of approximately -500mK. This amplitude is deeper than the standard prediction (in which the radio background is determined by the cosmic microwave background) by a factor of two and potentially hints at the existence of a radio background excess. While this result was obtained by fitting the data with a phenomenological flattened-Gaussian shape for the cosmological signal, here we develop a physical model for the inhomogeneous radio background sourced by the first galaxies hosting population III stars. Star formation in these galaxies is quenched at lower redshifts due to various feedback mechanisms, so they serve as a natural candidate for the excess radio background hinted by EDGES, without violating present day measurements by ARCADE2. We forward-model the EDGES sky temperature data, jointly sampling our physical model for the cosmic signal, a foreground model, and residual calibration errors. We compare the Bayesian evidences obtained by varying the complexity and prior ranges for the systematics. We find that the data is best explained by a model with seven log-polynomial foreground terms, and that it requires calibration residuals. Interestingly, the presence of a cosmic 21cm signal with a non-standard depth is decisively disfavored. This is contrary to previous EDGES analysis in the context of extra radio background models, serving as a caution against using a ''pseudo-likelihood'' built on a model (flattened Gaussian) that is different from the one being used for inference. We make our simulation code and associated emulator publicly-available.

astro-ph.CO↗

arXiv2024

Is ChatGPT Transforming Academics' Writing Style?

Based on one million arXiv papers submitted from May 2018 to January 2024, we assess the textual density of ChatGPT's writing style in their abstracts through a statistical analysis of word frequency changes. Our model is calibrated and validated on a mixture of real abstracts and ChatGPT-modified abstracts (simulated data) after a careful noise analysis. The words used for estimation are not fixed but adaptive, including those with decreasing frequency. We find that large language models (LLMs), represented by ChatGPT, are having an increasing impact on arXiv abstracts, especially in the field of computer science, where the fraction of LLM-style abstracts is estimated to be approximately 35%, if we take the responses of GPT-3.5 to one simple prompt, "revise the following sentences", as a baseline. We conclude with an analysis of both positive and negative aspects of the penetration of LLMs into academics' writing style.

cs.CL↗

arXiv2024

Are Large Language Models Chameleons? An Attempt to Simulate Social Surveys

Can large language models (LLMs) simulate social surveys? To answer this question, we conducted millions of simulations in which LLMs were asked to answer subjective questions. A comparison of different LLM responses with the European Social Survey (ESS) data suggests that the effect of prompts on bias and variability is fundamental, highlighting major cultural, age, and gender biases. We further discussed statistical methods for measuring the difference between LLM answers and survey data and proposed a novel measure inspired by Jaccard similarity, as LLM-generated responses are likely to have a smaller variance. Our experiments also reveal that it is important to analyze the robustness and variability of prompts before using LLMs to simulate social surveys, as their imitation abilities are approximate at best.

cs.CL↗

arXiv2024

21cmEMU: an emulator of 21cmFAST summary observables

Recent years have witnessed rapid progress in observations of the Epoch of Reionization (EoR). These have enabled high-dimensional inference of galaxy and intergalactic medium (IGM) properties during the first billion years of our Universe. However, even using efficient, semi-numerical simulations, traditional inference approaches that compute 3D lightcones on-the-fly can take $10^5$ core hours. Here 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: (i) the IGM mean neutral fraction; (ii) the 21-cm power spectrum; (iii) the mean 21-cm spin temperature; (iv) the sky-averaged (global) 21-cm signal; (v) the ultraviolet (UV) luminosity functions (LFs); and (vi) the Thomson scattering optical depth to the cosmic microwave background (CMB). All observables are predicted with sub-percent median accuracy, with a reduction of the computational cost by a factor of over 10$^4$. After validating inference results, we showcase a few applications, including: (i) quantifying the relative constraining power of different observational datasets; (ii) seeing how recent claims of a late EoR impact previous inferences; and (iii) forecasting upcoming constraints from the sixth observing season of the Hydrogen Epoch of Reionization Array (HERA) telescope. 21cmEMU is publicly-available, and is included as an alternative simulator in the public 21CMMC sampler.

astro-ph.CO↗

arXiv2024

Dark Matter Halo Parameters from Overheated Exoplanets via Bayesian Hierarchical Inference

Dark Matter (DM) can become captured, deposit annihilation energy, and hence increase the heat flow in exoplanets and brown dwarfs. Detecting such a DM-induced heating in a population of exoplanets in the inner kpc of the Milky Way thus provides potential sensitivity to the galactic DM halo parameters. We develop a Bayesian Hierarchical Model to investigate the feasibility of DM discovery with exoplanets and examine future prospects to recover the spatial distribution of DM in the Milky Way. We reconstruct from mock exoplanet datasets observable parameters such as exoplanet age, temperature, mass, and location, together with DM halo parameters, for representative choices of measurement uncertainty and the number of exoplanets detected. We find that detection of $\mathcal{O}(100)$ exoplanets in the inner Galaxy can yield quantitative information on the galactic DM density profile, under the assumption of 10% measurement uncertainty. Even as few as $\mathcal{O}(10)$ exoplanets can deliver meaningful sensitivities if the DM density and inner slope are sufficiently large.

astro-ph.IM↗

arXiv2024

SIDE-real: Supernova Ia Dust Extinction with truncated marginal neural ratio estimation applied to real data

We present the first fully simulation-based hierarchical analysis of the light curves of a population of low-redshift type Ia supernovae (SNae Ia). Our hardware-accelerated forward model, released in the Python package slicsim, includes stochastic variations of each SN's spectral flux distribution (based on the pre-trained BayeSN model), extinction from dust in the host and in the Milky Way, redshift, and realistic instrumental noise. By utilising truncated marginal neural ratio estimation (TMNRE), a neural network-enabled simulation-based inference technique, we implicitly marginalise over 4000 latent variables (for a set of $\approx 100$ SNae Ia) to efficiently infer SN Ia absolute magnitudes and host-galaxy dust properties at the population level while also constraining the parameters of individual objects. Amortisation of the inference procedure allows us to obtain coverage guarantees for our results through Bayesian validation and frequentist calibration. Furthermore, we show a detailed comparison to full likelihood-based inference, implemented through Hamiltonian Monte Carlo, on simulated data and then apply TMNRE to the light curves of 86 SNae Ia from the Carnegie Supernova Project, deriving marginal posteriors in excellent agreement with previous work. Given its ability to accommodate arbitrarily complex extensions to the forward model -- e.g. different populations based on host properties, redshift evolution, complicated photometric redshift estimates, selection effects, and non-Ia contamination -- without significant modifications to the inference procedure, TMNRE has the potential to become the tool of choice for cosmological parameter inference from future, large SN Ia samples.

astro-ph.CO↗

arXiv2024

Parameter estimation from Ly$α$ forest in Fourier space using Information Maximising Neural Network

We aim to present a robust parameter estimation with simulated Lya forest spectra from Sherwood-Relics simulations suite using Information Maximizing Neural Network(IMNN) to extract maximal information from Lya 1D-transmitted flux in Fourier space. We perform 1D estimations using IMNN for IGM thermal parameters $T_0$ & $γ$ at z=2-4 and cosmological parameters $σ_8$ & $n_s$ at z=3-4. We compare our results with estimates from power spectrum using posterior distribution from Markov Chain Monte Carlo(MCMC). We then check robustness of IMNN estimates against deviation in spectral noise levels,continuum uncertainties & instrumental smoothing effects. Using mock Lya forest sightlines from publicly available CAMELS project we also check the robustness of the trained IMNN on a different simulation. We also perform a 2D-parameter estimation for $T_0$ & HI photoionization rates $Γ_{HI}$. We obtain improved estimates of $T_0$ & $γ$ using IMNN over standard MCMC approach. These estimates are also more robust against SNR deviations at z=2 & 3. At z=4 the sensitivity to noise deviations is on par with MCMC estimates. The IMNN also provides $T_0$ and $γ$ estimates which are robust against continuum uncertainties by extracting continuum-independent small-scale information from Fourier domain. In case of $σ_8$ & $n_s$ IMNN performs on par with MCMC but still offers a significant speed boost in estimating parameters from a new dataset. The improved estimates with IMNN are seen for high instrumental-resolution(FWHM=6km/s). At medium or low resolutions IMNN performs similar to MCMC suggesting an improved extraction of small-scale information with IMNN. We also find that IMNN estimates are robust against the choice of simulation. By performing a 2D-parameter estimation for $T_0$ & $Γ_{HI}$ we also demonstrate how to take forward this approach observationally in the future.

astro-ph.CO↗

arXiv2024

Improved Weak Lensing Photometric Redshift Calibration via StratLearn and Hierarchical Modeling

Discrepancies between cosmological parameter estimates from cosmic shear surveys and from recent Planck cosmic microwave background measurements challenge the ability of the highly successful $Λ$CDM model to describe the nature of the Universe. To rule out systematic biases in cosmic shear survey analyses, accurate redshift calibration within tomographic bins is key. In this paper, we improve photo-$z$ calibration via Bayesian hierarchical modeling of full galaxy photo-$z$ conditional densities, by employing $\textit{StratLearn}$, a recently developed statistical methodology, which accounts for systematic differences in the distribution of the spectroscopic training/source set and the photometric target set. Using realistic simulations that were designed to resemble the KiDS+VIKING-450 dataset, we show that $\textit{StratLearn}$-estimated conditional densities improve the galaxy tomographic bin assignment, and that our $\textit{StratLearn}$-Bayesian framework leads to nearly unbiased estimates of the target population means. This leads to a factor of $\sim 2$ improvement upon the previously best photo-$z$ calibration method. Our approach delivers a maximum bias per tomographic bin of $Δ\langle z \rangle = 0.0095 \pm 0.0089$, with an average absolute bias of $0.0052 \pm 0.0067$ across the five tomographic bins.

astro-ph.CO↗

arXiv2023

SimSIMS: Simulation-based Supernova Ia Model Selection with thousands of latent variables

We present principled Bayesian model comparison through simulation-based neural classification applied to SN Ia analysis. We validate our approach on realistically simulated SN Ia light curve data, demonstrating its ability to recover posterior model probabilities while marginalizing over >4000 latent variables. The amortized nature of our technique allows us to explore the dependence of Bayes factors on the true parameters of simulated data, demonstrating Occam's razor for nested models. When applied to a sample of 86 low-redshift SNae Ia from the Carnegie Supernova Project, our method prefers a model with a single dust law and no magnitude step with host mass, disfavouring different dust laws for low- and high-mass hosts with odds in excess of 100:1.

astro-ph.CO↗

arXiv2023

Neural Posterior Estimation with guaranteed exact coverage: the ringdown of GW150914

We analyze the ringdown phase of the first detected black-hole merger, GW150914, using a simulation-based inference pipeline based on masked autoregressive flows. We obtain approximate marginal posterior distributions for the ringdown parameters, namely the mass, spin, and the amplitude and phases of the dominant mode and its first overtone. Thanks to the locally amortized nature of our method, we are able to calibrate our posteriors with injected simulations, producing posterior regions with guaranteed (i.e. exact) frequentist coverage of the true values. For GW150914, our calibrated posteriors provide only mild evidence (~ 2 sigma) for the presence of an overtone, even if the ringdown is assumed to start at the peak of the amplitude.

gr-qc↗

arXiv2023

Extreme data compression for Bayesian model comparison

We develop extreme data compression for use in Bayesian model comparison via the MOPED algorithm, as well as more general score compression. We find that Bayes factors from data compressed with the MOPED algorithm are identical to those from their uncompressed datasets when the models are linear and the errors Gaussian. In other nonlinear cases, whether nested or not, we find negligible differences in the Bayes factors, and show this explicitly for the Pantheon-SH0ES supernova dataset. We also investigate the sampling properties of the Bayesian Evidence as a frequentist statistic, and find that extreme data compression reduces the sampling variance of the Evidence, but has no impact on the sampling distribution of Bayes factors. Since model comparison can be a very computationally-intensive task, MOPED extreme data compression may present significant advantages in computational time.

astro-ph.IM↗

arXiv2023

Stratified Learning: A General-Purpose Statistical Method for Improved Learning under Covariate Shift

We propose a simple, statistically principled, and theoretically justified method to improve supervised learning when the training set is not representative, a situation known as covariate shift. We build upon a well-established methodology in causal inference, and show that the effects of covariate shift can be reduced or eliminated by conditioning on propensity scores. In practice, this is achieved by fitting learners within strata constructed by partitioning the data based on the estimated propensity scores, leading to approximately balanced covariates and much-improved target prediction. We demonstrate the effectiveness of our general-purpose method on two contemporary research questions in cosmology, outperforming state-of-the-art importance weighting methods. We obtain the best reported AUC (0.958) on the updated "Supernovae photometric classification challenge", and we improve upon existing conditional density estimation of galaxy redshift from Sloan Data Sky Survey (SDSS) data.

stat.ML↗

arXiv2023

SICRET: Supernova Ia Cosmology with truncated marginal neural Ratio EsTimation

Type Ia supernovae (SNae Ia), standardisable candles that allow tracing the expansion history of the Universe, are instrumental in constraining cosmological parameters, particularly dark energy. State-of-the-art likelihood-based analyses scale poorly to future large datasets, are limited to simplified probabilistic descriptions, and must explicitly sample a high-dimensional latent posterior to infer the few parameters of interest, which makes them inefficient. Marginal likelihood-free inference, on the other hand, is based on forward simulations of data, and thus can fully account for complicated redshift uncertainties, contamination from non-SN Ia sources, selection effects, and a realistic instrumental model. All latent parameters, including instrumental and survey-related ones, per-object and population-level properties, are implicitly marginalised, while the cosmological parameters of interest are inferred directly. As a proof of concept, we apply truncated marginal neural ratio estimation (TMNRE), a form of marginal likelihood-free inference, to BAHAMAS, a Bayesian hierarchical model for SALT parameters. We verify that TMNRE produces unbiased and precise posteriors for cosmological parameters from up to 100 000 SNae Ia. With minimal additional effort, we train a network to infer simultaneously the O(100 000) latent parameters of the supernovae (e.g. absolute brightnesses). In addition, we describe and apply a procedure that utilises local amortisation of the inference to convert the approximate Bayesian posteriors into frequentist confidence regions with exact coverage. Finally, we discuss the planned improvements to the model that are enabled by using a likelihood-free inference framework, like selection effects and non-Ia contamination.

astro-ph.CO↗

arXiv2022

Simple and statistically sound recommendations for analysing physical theories

Physical theories that depend on many parameters or are tested against data from many different experiments pose unique challenges to statistical inference. Many models in particle physics, astrophysics and cosmology fall into one or both of these categories. These issues are often sidestepped with statistically unsound ad hoc methods, involving intersection of parameter intervals estimated by multiple experiments, and random or grid sampling of model parameters. Whilst these methods are easy to apply, they exhibit pathologies even in low-dimensional parameter spaces, and quickly become problematic to use and interpret in higher dimensions. In this article we give clear guidance for going beyond these procedures, suggesting where possible simple methods for performing statistically sound inference, and recommendations of readily-available software tools and standards that can assist in doing so. Our aim is to provide any physicists lacking comprehensive statistical training with recommendations for reaching correct scientific conclusions, with only a modest increase in analysis burden. Our examples can be reproduced with the code publicly available at https://doi.org/10.5281/zenodo.4322283.

hep-ph↗

arXiv2021

Identification of high-energy astrophysical point sources via hierarchical Bayesian nonparametric clustering

The light we receive from distant astrophysical objects carries information about their origins and the physical mechanisms that power them. The study of these signals, however, is complicated by the fact that observations are often a mixture of the light emitted by multiple localized sources situated in a spatially-varying background. A general algorithm to achieve robust and accurate source identification in this case remains an open question in astrophysics. This paper focuses on high-energy light (such as X-rays and gamma-rays), for which observatories can detect individual photons (quanta of light), measuring their incoming direction, arrival time, and energy. Our proposed Bayesian methodology uses both the spatial and energy information to identify point sources, that is, separate them from the spatially-varying background, to estimate their number, and to compute the posterior probabilities that each photon originated from each identified source. This is accomplished via a Dirichlet process mixture while the background is simultaneously reconstructed via a flexible Bayesian nonparametric model based on B-splines. Our proposed method is validated with a suite of simulation studies and illustrated with an application to a complex region of the sky observed by the \emph{Fermi} Gamma-ray Space Telescope.

stat.AP↗

arXiv2020

Structure Formation Models Weaken Limits on WIMP Dark Matter from Dwarf Spheroidal Galaxies

Dwarf spheroidal galaxies that form in halo substructures provide stringent constraints on dark matter annihilation. Many ultrafaint dwarfs discovered with modern surveys contribute significantly to these constraints. At present, because of the lack of abundant stellar kinematic data for the ultrafaints, non-informative prior assumptions are usually made for the parameters of the density profiles. Based on semi-analytic models of dark matter subhalos and their connection to satellite galaxies, we present more informative and realistic satellite priors. We show that our satellite priors lead to constraints on the annihilation rate that are between a factor of 2 and a factor of 7 weaker than under non-informative priors. As a result, the thermal relic cross section can at best only be excluded (with 95% probability) for dark matter masses of $\lesssim 40$ GeV from dwarf spheroidal data, assuming annihilation into $b\bar{b}$.

astro-ph.CO↗