Related papers: Learning Theory Informed Priors for Bayesian Infer…
Modeling uncertainty in deep neural networks, despite recent important advances, is still an open problem. Bayesian neural networks are a powerful solution, where the prior over network weights is a design choice, often a normal…
Recently \citet{PEDE:2019ApJ} proposed a phenomenologically emergent dark energy (PEDE) which consider that the dark energy density evolves as $\widetilde{\Omega}_{\rm{DE}}(z)\,=\,\Omega_{\rm{DE,0}}\left[ 1 - {\rm{tanh}}\left(…
While many physics-based closure model forms have been posited for the sub-filter scale (SFS) in large eddy simulation (LES), vast amounts of data available from direct numerical simulation (DNS) create opportunities to leverage data-driven…
We develop a novel data-driven approach to the inverse problem of classical statistical mechanics: given experimental data on the collective motion of a classical many-body system, how does one characterise the free energy landscape of that…
Recently a full-shape analysis of large-scale structure (LSS) data was employed to provide new constraints on a class of Early Dark Energy (EDE) models. In this note, we derive similar constraints on New Early Dark Energy (NEDE) using the…
Recent advances in deep learning have led to a data-centric intelligence i.e. artificially intelligent models unlocking the potential to ingest a large amount of data and be really good at performing digital tasks such as text-to-image…
We propose a new cosmological model that considers dark matter as a barotropic fluid with a constant equation of state parameter and interprets dark energy as the phenomenological emergent dark energy rather than a cosmological constant.…
In the procedure of constraining the cosmological parameters with the observational Hubble data and the type Ia supernova data, the combination of Masked Autoregressive Flow and Denoising Autoencoder can perform a good result. The above…
Bayesian inference is known to provide a general framework for incorporating prior knowledge or specific properties into machine learning models via carefully choosing a prior distribution. In this work, we propose a new type of prior…
Early dark energy (EDE) is a proposed solution to the Hubble tension in which a new cosmological field accelerates cosmic expansion prior to recombination and reduces the physical size of the sound horizon. In previous work, a slight…
In Reinforcement Learning (RL), agents aim at maximizing cumulative rewards in a given environment. During the learning process, RL agents face the dilemma of exploitation and exploration: leveraging existing knowledge to acquire rewards or…
Recently, using Bayesian Machine Learning, a deviation from the cold dark matter model on cosmological scales has been put forward. Such model might replace a proposed non-gravitational interaction between dark energy and dark matter, and…
The data release six of the Atacama Cosmology Telescope (ACT DR6) and the second data release from the Dark Energy Spectroscopic Instrument (DESI DR2) recently became available. In light of these data, we update constraints on the Early…
This paper introduces a new neural network based prior for real valued functions on $\mathbb R^d$ which, by construction, is more easily and cheaply scaled up in the domain dimension $d$ compared to the usual Karhunen-Lo\`eve function space…
We consider the interplay of the Early Dark Energy (EDE) model, the Swampland Distance Conjecture (SDC), and cosmological parameter tensions. EDE is a proposed resolution of the Hubble tension relying upon a near-Planckian scalar field…
Early dark energy (EDE), introduced at the epoch of matter-radiation equality to alleviate the Hubble tension, has posed a new coincidence problem: why EDE appears at matter-radiation equality when their physics are completely unrelated? To…
This paper presents a comprehensive analysis of hyperparameter estimation within the empirical Bayes framework (EBF) for sparse learning. By studying the influence of hyperpriors on the solution of EBF, we establish a theoretical connection…
Diffusion models recently proved to be remarkable priors for Bayesian inverse problems. However, training these models typically requires access to large amounts of clean data, which could prove difficult in some settings. In this work, we…
Cosmological parameters and dark energy (DE) behavior are generally constrained assuming \textit{a priori} models. We work out a model-independent reconstruction to bound the key cosmological quantities and the DE evolution. Through the…
While energy-based models (EBMs) exhibit a number of desirable properties, training and sampling on high-dimensional datasets remains challenging. Inspired by recent progress on diffusion probabilistic models, we present a diffusion…