Explore the vast range of titles available.

Core-collapse supernovae - the deaths of massive stars - are among the most complex and diverse astrophysical phenomena, demonstrating a wide range of spectroscopic and photometric properties. These events result from the cessation of fusion in the cores of massive stars causing gravitational collapse, but there is a great deal that remains uncertain about the exact mechanisms involved. Studying the properties of populations of CCSNe can help constrain our knowledge of the physics involved in the explosion. In this thesis, I examine the properties of high-redshift core-collapse supernova in the Dark Energy Survey (DES) and compare with local samples from the Lick Observatory Supernova Search (LOSS) and Zwicky Transient Facility (ZTF). Comparing type II SNe in DES and ZTF, I see a difference in peak luminosity of 3.0σ significance and between LOSS and ZTF of 2.5σ. This could be caused by redshift evolution, although simpler causes such as differing levels of host galaxy extinction between the samples cannot be ruled out. I also examine host galaxy properties for these samples, finding an offset in host galaxy colour between DES and ZTF; for the same galaxy stellar mass, a DES galaxy is bluer than a ZTF galaxy. I consider a number of simple explanations for this - including galaxy evolution with redshift, selection biases in either the DES or ZTF samples, and systematic differences due to the different photometric bands available - but find that none can easily reconcile the differences in host colour between the two samples and thus its cause remains uncertain. During my analysis, I identified a very luminous SN IIb, DES14X2fna. This SN had an unusually high luminosity for its class, reaching ∼ −19.4 mag in r-band, and also declined rapidly after peak. SNe IIb are thought to be powered by the decay of 56Ni, but the mass of Ni that would be required to power this luminosity is inconsistent with the fast decline observed. This suggests that some other physics is involved. Using semi-analytic model fits, I show that 56Ni decay alone is unable to power this object, but interaction with surrounding circumstellar material (CSM) or the spin-down of a rapidly rotating neutron star formed in the explosion are two possible explanations for this unusual object. Finally, I explore the use of Generative Adversarial Networks (GANs) to generate synthetic CCSN light curves. GANs are a type of neural network used for data generation; this approach could be used to augment samples used to train photometric classification algorithms, improving their performance. By training on DES-like simulations I find that GANs are able to generate physically realistic light curves for a variety of CCSN types, demonstrating their potential to improve classification techniques going forward.

In recent years there has been significant debate around the impact of dust on SNe Ia, a major source of uncertainty in cosmological analyses. We perform the first cross-comparison of the probabilistic hierarchical SN Ia SED model BayeSN with the conventional SALT model, an important test given the history of conflicting conclusions regarding the distributions of host galaxy dust properties between the two. Applying BayeSN to SALT-based simulations, we find that BayeSN is able to accurately recover our simulated inputs, establishing excellent consistency between the two models. When inferring dust parameters with simulated samples including non-Ia contamination, we find that our choice of photometric classifier causes a bias in the inferred dust distribution; this arises because SNe Ia heavily impacted by dust are misclassified as contaminants and excluded. We then apply BayeSN to the sample of SNe from DES5YR to jointly infer host galaxy dust distributions and intrinsic differences on either side of the 'mass step' at \\(10^{10}\\) M\\(\\odot\\). We find evidence in favour of an intrinsic contribution to the mass step and differing \\(R_V\\) distributions. We also build on recent results supporting an environmental-dependence on the secondary maximum of SNe Ia in \\(i\\)-band. Twenty days post-peak, we find an offset in intrinsic \\(i\\)-band light curve between each mass bin at a significance in excess of \\(3\\sigma\\).

We present GausSN, a Bayesian semi-parametric Gaussian Process (GP) model for time-delay estimation with resolved systems of gravitationally lensed supernovae (glSNe). GausSN models the underlying light curve non-parametrically using a GP. Without assuming a template light curve for each SN type, GausSN fits for the time delays of all images using data in any number of wavelength filters simultaneously. We also introduce a novel time-varying magnification model to capture the effects of microlensing alongside time-delay estimation. In this analysis, we model the time-varying relative magnification as a sigmoid function, as well as a constant for comparison to existing time-delay estimation approaches. We demonstrate that GausSN provides robust time-delay estimates for simulations of glSNe from the Nancy Grace Roman Space Telescope and the Vera C. Rubin Observatory's Legacy Survey of Space and Time (Rubin-LSST). We find that up to 43.6% of time-delay estimates from Roman and 52.9% from Rubin-LSST have fractional errors of less than 5%. We then apply GausSN to SN Refsdal and find the time delay for the fifth image is consistent with the original analysis, regardless of microlensing treatment. Therefore, GausSN maintains the level of precision and accuracy achieved by existing time-delay extraction methods with fewer assumptions about the underlying shape of the light curve than template-based approaches, while incorporating microlensing into the statistical error budget. GausSN is scalable for time-delay cosmography analyses given current projections of glSNe discovery rates from Rubin-LSST and Roman.

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.

Type Ia supernovae (SNe Ia) are thermonuclear exploding stars that can be used to put constraints on the nature of our universe. One challenge with population analyses of SNe Ia is Malmquist bias, where we preferentially observe the brighter SNe due to limitations of our telescopes. If untreated, this bias can propagate through to our posteriors on cosmological parameters. In this paper, we develop a novel technique of using a normalising flow to learn the non-analytical likelihood of observing a SN Ia for a given survey from simulations, that is independent of any cosmological model. The learnt likelihood is then used in a hierarchical Bayesian model with Hamiltonian Monte Carlo sampling to put constraints on different sets of cosmological parameters conditioned on the observed data. We verify this technique on toy model simulations finding excellent agreement with analytically-derived posteriors to within \\(1 \\sigma\\).

Type Ia supernovae (SNe Ia) are standarizable candles whose observed light curves can be used to infer their distances, which can in turn be used in cosmological analyses. As the quantity of observed SNe Ia grows with current and upcoming surveys, increasingly scalable analyses are necessary to take full advantage of these new datasets for precise estimation of cosmological parameters. Bayesian inference methods enable fitting SN Ia light curves with robust uncertainty quantification, but traditional posterior sampling using Markov Chain Monte Carlo (MCMC) is computationally expensive. We present an implementation of variational inference (VI) to accelerate the fitting of SN Ia light curves using the BayeSN hierarchical Bayesian model for time-varying SN Ia spectral energy distributions (SEDs). We demonstrate and evaluate its performance on both simulated light curves and data from the Foundation Supernova Survey with two different forms of surrogate posterior -- a multivariate normal and a custom multivariate zero-lower-truncated normal distribution -- and compare them with the Laplace Approximation and full MCMC analysis. To validate of our variational approximation, we calculate the pareto-smoothed importance sampling (PSIS) diagnostic, and perform variational simulation-based calibration (VSBC). The VI approximation achieves similar results to MCMC but with an order-of-magnitude speedup for the inference of the photometric distance moduli. Overall, we show that VI is a promising method for scalable parameter inference that enables analysis of larger datasets for precision cosmology.

To reduce systematic uncertainties in Type Ia supernova (SN Ia) cosmology, the host galaxy dust law shape parameter, \\(R_V\\), must be accurately constrained. We thus develop a computationally-inexpensive pipeline, Bird-Snack, to rapidly infer dust population distributions from optical-near infrared SN colours at peak brightness, and determine which analysis choices significantly impact the population mean \\(R_V\\) inference, \\(\\mu_{R_V}\\). Our pipeline uses a 2D Gaussian process to measure peak \\(BVriJH\\) apparent magnitudes from SN light curves, and a hierarchical Bayesian model to simultaneously constrain population distributions of intrinsic and dust components. Fitting a low-to-moderate-reddening sample of 65 low-redshift SNe yields \\(\\mu_{R_V}=2.61^{+0.38}_{-0.35}\\), with \\(68\\%(95\\%)\\) posterior upper bounds on the population dispersion, \\(\\sigma_{R_V}<0.92(1.96)\\). This result is robust to various analysis choices, including: the model for intrinsic colour variations, fitting the shape hyperparameter of a gamma dust extinction distribution, and cutting the sample based on the availability of data near peak. However, these choices may be important if statistical uncertainties are reduced. With larger near-future optical and near-infrared SN samples, Bird-Snack can be used to better constrain dust distributions, and investigate potential correlations with host galaxy properties. Bird-Snack is publicly available; the modular infrastructure facilitates rapid exploration of custom analysis choices, and quick fits to simulated datasets, for better interpretation of real-data inferences.

We apply the hierarchical probabilistic SED model BayeSN to analyse a sample of 475 SNe Ia (0.015 < z < 0.4) from Foundation, DES3YR and PS1MD to investigate the properties of dust in their host galaxies. We jointly infer the dust law \\(R_V\\) population distributions at the SED level in high- and low-mass galaxies simultaneously with dust-independent, intrinsic differences. We find an intrinsic mass step of \\(-0.049\\pm0.016\\) mag, at a significance of 3.1\\(\\sigma\\), when allowing for a constant intrinsic, achromatic magnitude offset. We additionally apply a model allowing for time- and wavelength-dependent intrinsic differences between SNe Ia in different mass bins, finding \\(\\sim\\)2\\(\\sigma\\) differences in magnitude and colour around peak and 4.5\\(\\sigma\\) differences at later times. These intrinsic differences are inferred simultaneously with a difference in population mean \\(R_V\\) of \\(\\sim\\)2\\(\\sigma\\) significance, demonstrating that both intrinsic and extrinsic differences may play a role in causing the host galaxy mass step. We also consider a model which allows the mean of the \\(R_V\\) distribution to linearly evolve with redshift but find no evidence for any evolution - we infer the gradient of this relation \\(\\eta_R = -0.38\\pm0.70\\). In addition, we discuss in brief a new, GPU-accelerated Python implementation of BayeSN suitable for application to large surveys which is publicly available and can be used for future cosmological analyses; this code can be found here: https://github.com/bayesn/bayesn.

We use hierarchical Bayesian modelling to calibrate a network of 32 all-sky faint DA white dwarf (DA WD) spectrophotometric standards (\\(16.5 < V < 19.5\\)) alongside the three CALSPEC standards, from 912 Å to 32 \\(\\mu\\)m. The framework is the first of its kind to jointly infer photometric zeropoints and WD parameters (\\(\\log g\\), \\(T_{\\text{eff}}\\), \\(A_V\\), \\(R_V\\)) by simultaneously modelling both photometric and spectroscopic data. We model panchromatic HST/WFC3 UVIS and IR fluxes, HST/STIS UV spectroscopy and ground-based optical spectroscopy to sub-percent precision. Photometric residuals for the sample are the lowest yet yielding \\(<0.004\\) mag RMS on average from the UV to the NIR, achieved by jointly inferring time-dependent changes in system sensitivity and WFC3/IR count-rate nonlinearity. Our GPU-accelerated implementation enables efficient sampling via Hamiltonian Monte Carlo, critical for exploring the high-dimensional posterior space. The hierarchical nature of the model enables population analysis of intrinsic WD and dust parameters. Inferred SEDs from this model will be essential for calibrating the James Webb Space Telescope as well as next-generation surveys, including Vera Rubin Observatory's Legacy Survey of Space and Time, and the Nancy Grace Roman Space Telescope.

We present the MULTIMODAL UNIVERSE, a large-scale multimodal dataset of scientific astronomical data, compiled specifically to facilitate machine learning research. Overall, the MULTIMODAL UNIVERSE contains hundreds of millions of astronomical observations, constituting 100\\,TB of multi-channel and hyper-spectral images, spectra, multivariate time series, as well as a wide variety of associated scientific measurements and \"metadata\". In addition, we include a range of benchmark tasks representative of standard practices for machine learning methods in astrophysics. This massive dataset will enable the development of large multi-modal models specifically targeted towards scientific applications. All codes used to compile the MULTIMODAL UNIVERSE and a description of how to access the data is available at https://github.com/MultimodalUniverse/MultimodalUniverse