Related papers: PINION: Physics-informed neural network for accele…
Message passing neural networks have become a method of choice for learning on graphs, in particular the prediction of chemical properties and the acceleration of molecular dynamics studies. While they readily scale to large training data…
The ability of the future low frequency component of the Square Kilometre Array radio telescope (SKA-Low) to produce tomographic images of the redshifted 21-cm signal will enable direct studies of the evolution of the sizes and shapes of…
This research presents the development of an innovative algorithm tailored for the adaptive sampling of residual points within the framework of Physics-Informed Neural Networks (PINNs). By addressing the limitations inherent in existing…
We present PIVONet (Physically-Informed Variational ODE Neural Network), a unified framework that integrates Neural Ordinary Differential Equations (Neuro-ODEs) with Continuous Normalizing Flows (CNFs) for stochastic fluid simulation and…
We produce simulations of emission of the atomic CII line in large sky fields in order to determine the current prospects for mapping this line during the high redshift Epoch of Reionization. We estimate the CII line intensity, redshift…
We carry out hydrodynamical simulations of galaxy formation that simultaneously follow radiative transfer of hydrogen-ionizing photons, based on the optically-thin variable Eddington tensor approximation as implemented in the {\small…
Three-dimensional radiative transfer simulations of the epoch of reionization can produce realistic results, but are computationally expensive. On the other hand, simulations relying on one-dimensional radiative transfer solutions are…
Estimating rate coefficients from complex chemical reactions is essential for advancing detailed chemistry. However, the stiffness inherent in real-world atmospheric chemistry systems poses severe challenges, leading to training instability…
We introduce the THESAN project, a suite of large volume (L = 95.5 cMpc) radiation-magnetohydrodynamic simulations that simultaneously model the large-scale statistical properties of the IGM during reionization and the resolved…
Phase-field equations, mostly solved numerically, are known for capturing the mesoscale microstructural evolution of a material. However, such numerical solvers are computationally expensive as it needs to generate fine mesh systems to…
We introduce NeuroPINNs, a neuroscience-inspired extension of Physics-Informed Neural Networks (PINNs) that incorporates biologically motivated spiking neuron models to achieve energy-efficient PDE solving. Unlike conventional PINNs, which…
The local intensity of the 21 cm signal emitted during the Epoch of Reionization that will be mapped by the SKA is modulated by the amount of neutral hydrogen. Consequently, understanding the process of reionization of the intergalactic…
Ultrafast optics is driven by a myriad of complex nonlinear dynamics. The ubiquitous presence of governing equations in the form of partial integro-differential equations (PIDE) necessitates the need for advanced computational tools to…
Physics-informed neural networks (PINNs) offer a promising framework by embedding partial differential equations (PDEs) into the loss function together with measurement data, making them well-suited for inverse problems. However, standard…
The designs of the first generation of cosmological 21-cm observatories are split between single dipole experiments which integrate over a large patch of sky in order to find the global (spectral) signature of reionization, and…
The simulation of power system dynamics poses a computationally expensive task. Considering the growing uncertainty of generation and demand patterns, thousands of scenarios need to be continuously assessed to ensure the safety of power…
In this first of several application papers, we investigate the mechanics of reionization from stellar sources in high-z galaxies, the utility of various clumping factors on estimating the recombination time in the IGM, and the photon…
The process of heating and reionization of the Universe at high redshift links small scale structures/galaxy formation and large scale inter-galactic medium properties. Even if the first one is difficult to observe, an observation window is…
We develop improved physics-informed neural networks (PINNs) for high-order and high-dimensional power system models described by nonlinear ordinary differential equations. We propose some novel enhancements to improve PINN training and…
We propose a deep learning analyzing technique with convolutional neural network (CNN) to predict the evolutionary track of the Epoch of Reionization (EoR) from the 21-cm differential brightness temperature tomography images. We use…