Related papers: Learning Theory Informed Priors for Bayesian Infer…

We review and update constraints on the Early Dark Energy (EDE) model from cosmological data sets, in particular Planck PR3 and PR4 cosmic microwave background (CMB) data and large-scale structure (LSS) data sets including galaxy clustering…

Physics-informed extreme learning machine (PIELM) has recently received significant attention as a rapid version of physics-informed neural network (PINN) for solving partial differential equations (PDEs). The key characteristic is to fix…

We investigate the use of approximate Bayesian neural networks (BNNs) in modeling hundreds of time-delay gravitational lenses for Hubble constant ($H_0$) determination. Our BNN was trained on synthetic HST-quality images of strongly lensed…

In this paper, we investigate the process in which axion dark matter undergoes thermal friction, resulting in energy injection into dark radiation, with the aim of mitigating the Hubble tension and large-scale structure tension. In the…

Data scarcity, bias, and experimental noise are all frequently encountered problems in the application of deep learning to chemical and material science disciplines. Transfer learning has proven effective in compensating for the lack in…

Bayesian neural networks (BNNs) have recently gained popularity due to their ability to quantify model uncertainty. However, specifying a prior for BNNs that captures relevant domain knowledge is often extremely challenging. In this work,…

We introduce a computational efficient data-driven framework suitable for quantifying the uncertainty in physical parameters and model formulation of computer models, represented by differential equations. We construct physics-informed…

Differences in the values of the Hubble constant obtained from the local universe and the early universe have resulted in a significant tension. This tension signifies that our understanding of cosmology (physical processes and/or…

In light of recent observations by the Dark Energy Spectroscopic Instrument (DESI), we study evidence for thawing quintessence over a cosmological constant as dark energy, with emphasis on the effect of the choice of priors. Working with a…

Recently the Dark Energy Spectroscopic Instrument (DESI) provided constraints on the expansion history from their Data Release 2 (DR2). The DESI baryon acoustic oscillation (BAO) measurements are well described by a flat $\Lambda$CDM model,…

One of the core facets of Bayesianism is in the updating of prior beliefs in light of new evidence$\text{ -- }$so how can we maintain a Bayesian approach if we have no prior beliefs in the first place? This is one of the central challenges…

We propose a family of novel hierarchical Bayesian deep auto-encoder models capable of identifying disentangled factors of variability in data. While many recent attempts at factor disentanglement have focused on sophisticated learning…

Learning probabilistic models that can estimate the density of a given set of samples, and generate samples from that density, is one of the fundamental challenges in unsupervised machine learning. We introduce a new generative model based…

Among likelihood-based approaches for deep generative modelling, variational autoencoders (VAEs) offer scalable amortized posterior inference and fast sampling. However, VAEs are also more and more outperformed by competing models such as…

We update the distance priors by adopting $Planck~ \textrm{TT,TE,EE}+\textrm{lowP}$ data released in 2015, and our results impose at least $30\%$ tighter constraints than those from $Planck~ \textrm{TT}+\textrm{lowP}$. Combining the…

We introduce a unified model of early and late dark energy. We call it {\it quintessential early dark energy} model where early and late dark energy are explained by a single scalar field {\it i.e.}, two different energy scales are related…

We study the fit of cosmological models with two additional free parameters $N_\mathrm{eff}$ and $\xi_e$ in addition to the parameters of $\Lambda$CDM. We introduce extra radiation components such as hot axions or sterile neutrinos in…

We revisit a cosmological model where dark matter (DM) and dark energy (DE) follow barotropic equations of state, allowing deviations from the standard $\Lambda$CDM framework (i.e. $w_{dm} \neq 0$, $w_{de} \neq -1$), considering both flat…

Constraining the Dark Energy (DE) equation of state, w, is one of the primary science goals of ongoing and future cosmological surveys. In practice, with imperfect data and incomplete redshift coverage, this requires making assumptions…

Physics-informed neural networks (PINNs) provide a mesh-free framework for solving PDE-constrained inverse problems, but their extension to Bayesian inversion still faces a fundamental difficulty: prior distributions are typically defined…