Browsing by Author "Hassan, Sultan"

Now showing 1 - 8 of 8
  • Thumbnail Image
    Item
    The CAMELS multifield data set: Learning the universe’s fundamental parameters with artificial intelligence
    (IOP, 2022) Shy, Genel; Villaescusa-Navarro, Fransisco; Anglés-Alcázar, Daniel; Dave, Romeel; Hassan, Sultan
    We present the Cosmology and Astrophysics with Machine Learning Simulations (CAMELS) Multifield Data set (CMD), a collection of hundreds of thousands of 2D maps and 3D grids containing many different properties of cosmic gas, dark matter, and stars from more than 2000 distinct simulated universes at several cosmic times. The 2D maps and 3D grids represent cosmic regions that span ∼100 million light-years and have been generated from thousands of state-of-the-art hydrodynamic and gravity-only N-body simulations from the CAMELS project. Designed to train machine-learning models, CMD is the largest data set of its kind containing more than 70 TB of data. In this paper we describe CMD in detail and outline a few of its applications. We focus our attention on one such task, parameter inference, formulating the problems we face as a challenge to the community.
  • Thumbnail Image
    Item
    Constraining the astrophysics and cosmology from 21 cm tomography using deep learning with the SKA
    (Oxford University Press, 2020) Hassan, Sultan; Andrianomena, Sambatra; Doughty, Caitlin
    Future Square Kilometre Array (SKA) surveys are expected to generate huge data sets of 21 cm maps on cosmological scales from the Epoch of Reionization. We assess the viability of exploiting machine learning techniques, namely, convolutional neural networks (CNNs), to simultaneously estimate the astrophysical and cosmological parameters from 21 cm maps from seminumerical simulations. We further convert the simulated 21 cm maps into SKA-like mock maps using the detailed SKA antennae distribution, thermal noise, and a recipe for foreground cleaning. We successfully design two CNN architectures (VGGNet-like and ResNet-like) that are both efficiently able to extract simultaneously three astrophysical parameters, namely the photon escape fraction (fesc), the ionizing emissivity power dependence on halo mass (Cion), and the ionizing emissivity redshift evolution index (Dion), and three cosmological parameters, namely the matter density parameter (Ωm), the dimensionless Hubble constant (h), and the matter fluctuation amplitude (σ8), from 21 cm maps at several redshifts.
  • Thumbnail Image
    Item
    Constraining the reionization history using deep learning from 21-cm tomography with the Square Kilometre Array
    (Oxford University Press, 2020) Mangena, Tumelo; Hassan, Sultan; Santos, Mario G.
    Upcoming 21-cm surveys with the SKA1-LOW telescope will enable imaging of the neutral hydrogen distribution on cosmological scales in the early Universe. These surveys are expected to generate huge imaging data sets that will encode more information than the power spectrum. This provides an alternative unique way to constrain the reionization history, which might break the degeneracy in the power spectral analysis. Using convolutional neural networks, we create a fast estimator of the neutral fraction from the 21-cm maps that are produced by our large seminumerical simulation. Our estimator is able to efficiently recover the neutral fraction (⁠xHI⁠) at several redshifts with a high accuracy of 99 per cent as quantified by the coefficient of determination R2.
  • Thumbnail Image
    Item
    Deep Learning Voigt Profiles. I. Single-Cloud Doublets
    (2024) Stemock, Bryson; Hassan, Sultan; Churchill, Christopher W
    Voigt profile (VP) decomposition of quasar absorption lines is key to studying intergalactic gas and the baryon cycle governing the formation and evolution of galaxies. The VP velocities, column densities, and Doppler b parameters inform us of the kinematic, chemical, and ionization conditions of these astrophysical environments. A drawback of traditional VP fitting is that it can be human-time intensive. With the coming next generation of large all-sky survey telescopes with multi object high-resolution spectrographs, the time demands will significantly outstrip our resources. Deep learning pipelines hold the promise to keep pace and deliver science-digestible data products. We explore the application of deep learning convolutional neural networks (CNNs) for predicting VP-fitted parameters directly from the normalized pixel flux values in quasar absorption line profiles. A CNN was applied to 56 single-component Mg II λλ2796, 2803 doublet absorption line systems observed with HIRES and UVES (R = 45,000). The CNN predictions were statistically indistinct from those of a traditional VP fitter. The advantage is that, once trained, the CNN processes systems ∼105 times faster than a human expert fitting VP profiles by hand. Our pilot study shows that CNNs hold promise to perform bulk analysis of quasar absorption line systems in the future.
  • Thumbnail Image
    Item
    Hiflow: Generating diverse hi maps and inferring cosmology while marginalizing over astrophysics using normalizing flows
    (IOP Publishing, 2022) Hassan, Sultan; Villaescusa-Navarro, Francisco; Wandelt, Benjamin
    A wealth of cosmological and astrophysical information is expected from many ongoing and upcoming large-scale surveys. It is crucial to prepare for these surveys now and develop tools that can efficiently extract most information. We present HIFLOW: a fast generative model of the neutral hydrogen (HI) maps that is conditioned only on cosmology (Ωm and σ8) and designed using a class of normalizing flow models, the masked autoregressive flow. HIFLOW is trained on the state-of-the-art simulations from the Cosmology and Astrophysics with MachinE Learning Simulations (CAMELS) project. HIFLOW has the ability to generate realistic diverse maps without explicitly incorporating the expected two-dimensional maps structure into the flow as an inductive bias. We find that HIFLOW is able to reproduce the CAMELS average and standard deviation HI power spectrum within a factor of 2, scoring a very high R2 > 90%. By inverting the flow, HIFLOW provides a tractable high-dimensional likelihood for efficient parameter inference. We show that the conditional HIFLOW on cosmology is successfully able to marginalize over astrophysics at the field level, regardless of the stellar and AGN feedback strengths. This new tool represents a first step toward a more powerful parameter inference, maximizing the scientific return of future HI surveys, and opening a new avenue to minimize the loss of complex information due to data compression down to summary statistics.
  • Thumbnail Image
    Item
    Reconstruction of the ionization history from 21cm maps with deep learning
    (University of the Western Cape, 2020) Mangena; Santos, Mario; Hassan, Sultan
    Upcoming and ongoing 21cm surveys, such as the Square Kilometre Array (SKA), Hydrogen Epoch of Reionization Array (HERA) and Low Frequency Array (LOFAR), will enable imaging of the neutral hydrogen distribution on cosmological scales in the early Universe. These experiments are expected to generate huge imaging datasets that will encode more information than the power spectrum. This provides an alternative unique way to constrain the astrophysical and cosmological parameters, which might break the degeneracies in the power spectral analysis. The global history of reionization remains fairly unconstrained. In this thesis, we explore the viability of directly using the 21cm images to reconstruct and constrain the reionization history. Using Convolutional Neural Networks (CNN), we create a fast estimator of the global ionization fraction from the 21cm images as produced by our Large Semi-numerical Simulation (SimFast21). Our estimator is able to efficiently recover the ionization fraction (xHII) at several redshifts, z = 7; 8; 9; 10 with an accuracy of 99% as quantified by the coefficient of determination R2 without being given any additional information about the 21cm maps. This approach, contrary to estimations based on the power spectrum, is model independent. When adding the thermal noise and instrumental effects from these 21cm arrays, the results are sensitive to the foreground removal level, affecting the recovery of high neutral fractions. We also observe similar trend when combining all redshifts but with an improved accuracy. Our analysis can be easily extended to place additional constraints on other astrophysical parameters such as the photon escape fraction. This work represents a step forward to extract the astrophysical and cosmological information from upcoming 21cm surveys.
  • Thumbnail Image
    Item
    Simulating the neutral hydrogen distribution during cosmic reionization
    (University of the Western Cape, 2018) Hassan, Sultan; Dave, Romeel
    We improve on the physical treatment of ionising source and sink populations in the large scale semi-numerical simulations by implementing new physically motivated parametrizations taken from high-resolution radiative transfer simulations, in order to account for the non-linear dependence on halo mass, redshift and environment. This provides an efficient unique way to connect the small scale astrophysics to the large scale cosmology. These new parametrizations allow the model to simultaneously match all current reionization observations with only 4% photon escape fraction. These improvements result in 2-3 x 21cm power spectrum variations on small and large scales, and hence showing the importance of accurately treating ionising sources and sinks in 21cm simulations. We further implement time-integrated effects to accurately track the evolution of ionising photons, inhomogeneous recombinations and partially ionized regions during reionization. Including these effects yields larger HII regions and a more sudden reionization, which leads to an order of magnitude more 21cm power on large scales. We develop a robust parameter estimation pipeline to constrain the model astrophysical parametersagainst several reionization observations. We find that future 21cm observations provide tighter constraints on the astrophysical parameters and complement different derived constraints from other reionization observations. We finally employ the high redshift observations to add ionising photons from Active Galactic Nuclei (AGN), in order to assess the ability of AGN-dominated models to solely complete reionization. Unlike the case with galaxies, the AGN-only models cannot simultaneously match all current reionization observations. AGN-only models produce 21cm power spectrum that is 2 x higher on all scales as compared with galaxies-dominated models. Future 21cm surveys will play a key role to distinguishing between these two scenarios, even though AGN are highly unlikely to drive cosmic reionization.
  • Thumbnail Image
    Item
    Testing galaxy formation simulations with damped Lyman-α abundance and metallicity evolution
    (Oxford University Press, 2020) Hassan, Sultan; Finlator, Kristian; Dave, Romeel
    We examine the properties of damped Lyman-α absorbers (DLAs) emerging from a single set of cosmological initial conditions in two state-of-the-art cosmological hydrodynamic simulations: SIMBA and TECHNICOLOR DAWN. The former includes star formation and black hole feedback treatments that yield a good match with low-redshift galaxy properties, while the latter uses multifrequency radiative transfer to model an inhomogeneous ultraviolet background (UVB) self-consistently and is calibrated to match the Thomson scattering optical depth, UVB amplitude, and Ly α forest mean transmission at z > 5. Both simulations are in reasonable agreement with the measured stellar mass and star formation rate functions at z ≥ 3, and both reproduce the observed neutral hydrogen cosmological mass density, ΩHI(z)⁠.