[PDF][PDF] Kinetic. jl: A portable finite volume toolbox for scientific and neural computing
T Xiao - Journal of Open Source Software, 2021 - joss.theoj.org
Kinetic. jl is a lightweight finite volume toolbox written in the Julia programming language for
the study of computational physics and scientific machine learning. It is an open-source …
the study of computational physics and scientific machine learning. It is an open-source …
[PDF][PDF] AN IMPLEMENTATION OF NEURAL ORDINARY DIFFERENTIAL EQUATIONS
R ARRIETA - raw.githubusercontent.com
1. Introduction. In this report a review of Neural Ordinary Differential Equations (Neural
ODEs), as well as an implementation coded in Julia, is presented. Neural ODEs are a class of …
ODEs), as well as an implementation coded in Julia, is presented. Neural ODEs are a class of …
Machine learning accelerated particle-in-cell plasma simulations
Particle-In-Cell (PIC) methods are frequently used for kinetic, high-fidelity simulations of
plasmas. Implicit formulations of PIC algorithms feature strong conservation properties, up to …
plasmas. Implicit formulations of PIC algorithms feature strong conservation properties, up to …
Opening the blackbox: Accelerating neural differential equations by regularizing internal solver heuristics
… 2018). We note that discrete adjoints are known to be more stable than continuous adjoints
(Zhang & Sandu, 2014) and in the context of neural ODEs have been shown to stabilize the …
(Zhang & Sandu, 2014) and in the context of neural ODEs have been shown to stabilize the …
[PDF][PDF] Deep Networks for Optical Phase Retrieval
H Swan - cs230.stanford.edu
Modern tools in optics allow for generation of laser beams with essentially arbitrary intensity
patterns. However, in practice doing so requires solution of a computationally intensive …
patterns. However, in practice doing so requires solution of a computationally intensive …
Dicrotic notch identification: A generalizable hybrid approach under arterial blood pressure (ABP) curve deformations
M Saffarpour, D Basu, F Radaei, K Vali… - 2021 43rd Annual …, 2021 - ieeexplore.ieee.org
Dicrotic Notch (DN) is a distinctive and clinically significant feature of the arterial blood
pressure curve. Its automatic identification has been the focus of many kinds of research using …
pressure curve. Its automatic identification has been the focus of many kinds of research using …
Raytracer. jl: A differentiable renderer that supports parameter optimization for scene reconstruction
A Pal - arXiv preprint arXiv:1907.07198, 2019 - arxiv.org
In this paper, we present RayTracer.jl, a renderer in Julia that is fully differentiable using
source-to-source Automatic Differentiation (AD). This means that RayTracer not only renders 2D …
source-to-source Automatic Differentiation (AD). This means that RayTracer not only renders 2D …
Enhancing the interoperability between deep learning frameworks by model conversion
Y Liu, C Chen, R Zhang, T Qin, X Ji, H Lin… - Proceedings of the 28th …, 2020 - dl.acm.org
… The core task is to pre-identify the target nodes whose operators are non-transpose-commutative …
2018. A general reinforcement learning algorithm that masters chess, shogi, and Go …
2018. A general reinforcement learning algorithm that masters chess, shogi, and Go …
Estimating invariant sets using physics-informed neural networks
VV Tadiparthi, R Bhattacharya - AIAA SCITECH 2022 Forum, 2022 - arc.aiaa.org
View Video Presentation: https://doi.org/10.2514/6.2022-1441.vid In this paper, we seek to
estimate the invariant sets of stochastic dynamical systems by solving the steady-state form of …
estimate the invariant sets of stochastic dynamical systems by solving the steady-state form of …