Burgers' equation
PulseAugur coverage of Burgers' equation — every cluster mentioning Burgers' equation across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
New DEFT method boosts efficiency in modeling complex physical systems
Researchers have introduced DEFT, a novel data-efficient sampling method for modeling spatiotemporal dynamical systems governed by partial differential equations. This frequency-domain approach identifies dominant Fouri…
-
New theory quantifies neural operator approximation in Sobolev spaces
Researchers have developed a new theoretical framework for understanding neural operators' approximation capabilities within Sobolev spaces. This framework establishes an explicit relationship between model complexity a…
-
Structure-Preserving Neural Networks Enhance Burgers' Equation Simulations
Researchers have developed a novel machine learning method for creating subgrid-scale (SGS) parametrizations in coarse simulations of partial differential equations. This approach utilizes structure-preserving neural ne…
-
IG-GAN uses intrinsic geometry for aerodynamic data generation
Researchers have developed IG-GAN, a novel generative adversarial network designed to handle data that exists on manifolds rather than flat Euclidean space, a common characteristic of real-world data, particularly in ae…
-
SPARC-Net architecture improves physics-informed neural networks for complex PDEs
Researchers have developed SPARC-Net, a novel architecture and training framework designed to overcome limitations in Physics-Informed Neural Networks (PINNs) when solving complex partial differential equations (PDEs). …
-
New SPARC-Net architecture enhances physics-informed neural networks for complex PDEs
Researchers have developed SPARC-Net, a novel architecture designed to overcome limitations in physics-informed neural networks (PINNs) when dealing with stiff and shock-dominated partial differential equations (PDEs). …
-
New framework boosts physics-informed neural network accuracy
Researchers have developed DSGNAR, a novel optimization framework designed to improve the training of physics-informed neural networks (PINNs). This framework addresses the ill-conditioning issues that have previously l…
-
New deep learning framework MuRFiV enhances spatiotemporal dynamics prediction
Researchers have developed a new deep learning framework called MuRFiV, inspired by finite-volume methods, to improve the prediction of complex spatiotemporal dynamics. This framework integrates physics-informed learnin…
-
New Math Model for Neural Network Initialization Spectra
Researchers have developed a new mathematical framework to analyze the singular value spectrum of products of non-square random matrices. This framework is applicable to understanding the feature covariance eigenvalues …
-
Hartley Neural Operator offers real-valued alternative to Fourier Neural Operators
Researchers have introduced the Hartley Neural Operator (HNO), a new model designed to mirror the capabilities of Fourier Neural Operators (FNO) but with a focus on real-valued partial differential equation (PDE) soluti…
-
New neural networks accelerate PDE solving with improved accuracy and speed · 4 sources tracked
Researchers are developing advanced neural network architectures to improve the solving of partial differential equations (PDEs). One approach, Adaptive Hard-Soft Physics-Informed Neural Networks (HSPINN), enforces boun…
-
ANCHOR framework enhances neural operator accuracy for PDE simulations
Researchers have developed ANCHOR, a novel framework that combines neural operators with classical numerical solvers to improve the accuracy and stability of simulating time-dependent partial differential equations (PDE…
-
New deep learning method tackles complex PDE optimization
Researchers have developed a novel two-stage multi-grade deep learning (TS-MGDL) method to address the optimization challenges in training deep neural networks for partial differential equations (PDEs). This approach fi…
-
New Adaptive Memory Gate Enhances Neural Operator Performance
Researchers have developed an Adaptive Memory Gate for Neural Operators (AMGFNO) to improve their performance in solving time-dependent partial differential equations (PDEs). Existing memory-augmented neural operators u…
-
New active learning method discovers dynamics with ultra-low data
Researchers have developed a new active learning strategy to discover the governing equations of complex dynamical systems, particularly in scenarios where data is scarce. This method, building on Sparse Identification …
-
New Wavelet-Laplace Neural Operator Enhances PDE Solving
Researchers have introduced the Wavelet-Laplace Neural Operator (WLNO), a new neural operator designed to solve partial differential equations. WLNO enhances the existing Laplace Neural Operator (LNO) by incorporating a…
-
Deep Neural Networks viewed as Discrete Dynamical Systems
A new research paper proposes viewing deep neural networks (DNNs) as discrete dynamical systems, drawing parallels to neural integral equations and their PDE forms. The study compares numerical solutions of Burgers' and…
-
New neural operator integrates physics symmetries for improved generalization
Researchers have developed a new neural operator called PACE-FNO that better handles out-of-distribution scenarios by incorporating known continuous symmetries of evolution equations. This model separates the tasks of e…
-
New kernel learning method tackles nonlinear PDEs with multifidelity data
Researchers have developed a new kernel learning approach using cokriging to solve nonlinear partial differential equations (PDEs). This method leverages empirical information from multifidelity simulations to fit a dif…
-
New Pi-PINN framework enhances physics-informed neural network generalization
Researchers have developed a new framework called Pi-PINN to improve the generalization capabilities of physics-informed neural networks (PINNs). This approach learns transferable physics-informed representations, allow…