partial differential equations
PulseAugur coverage of partial differential equations — every cluster mentioning partial differential equations across labs, papers, and developer communities, ranked by signal.
- instance of alphaXiv 90%
- used by Neural Operators 90%
- used by alphaXiv 70%
- used by ScienceCast 70%
- used by Gotit.pub 70%
- instance of Fourier Neural Operators 70%
- used by Influence Flower 70%
- instance of Gotit.pub 70%
- used by Fourier Neural Operators 70%
- instance of Neural Operators 70%
- used by finite element method 70%
- competes with Fourier Neural Operators 60%
14 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 neural operators tackle PDEs with oscillator dynamics and optimal transport
Two new research papers introduce novel neural operator architectures for solving partial differential equations (PDEs). The Kuramoto Neural Operator (KNO) leverages coupled oscillator dynamics to represent solutions, s…
-
ProPINN architecture tackles propagation failures in physics-informed neural networks
Researchers have introduced ProPINN, a novel architecture designed to address propagation failures in physics-informed neural networks (PINNs). These failures occur when supervision signals from initial or boundary cond…
-
New PINN training methods tackle high-frequency and parameterized PDEs · 4 sources tracked
Researchers have developed new methods to improve the training of physics-informed neural networks (PINNs), addressing challenges like spectral bias and representation-coefficient coupling. One approach, IFeF-PINN, uses…
-
LLMs integrated into partial differential equation workflows, paper finds
A new arXiv paper explores the integration of large language models (LLMs) into workflows for solving partial differential equations (PDEs). The paper details how LLMs can assist in formulating governing models, generat…
-
Barron Spaces Enable Neural Network Approximation of High-Dimensional PDEs
Researchers have developed a method to approximate solutions for high-dimensional second-order elliptic partial differential equations (PDEs) using Barron spaces. This approach demonstrates that such solutions can be ap…
-
New neural operator frameworks tackle complex partial differential equations · 2 papers
Two new research papers introduce novel neural operator frameworks for solving partial differential equations (PDEs). The first, FB-C2CNet, utilizes fixed bases to encode and decode function coefficients, reducing train…
-
New Feature Interaction Models Enhance Physics-Informed Neural Networks
Researchers have developed new methods to enhance the expressiveness of physics-informed neural networks (PINNs) and neural operators. By incorporating feature interaction modules inspired by factorization machines, the…
-
New HERO training method boosts long-horizon accuracy for neural operators
Researchers have introduced HERO (History-Enriched Rollout Training), a novel method designed to improve the long-horizon accuracy of autoregressive neural operators. This technique addresses the issue of error accumula…
-
AI agents discover physics mappings and new algorithms for neural networks · 2 sources tracked
Two new research papers explore the use of AI agents in scientific discovery, specifically within physics and computational mathematics. The first paper introduces StatMechBench-v0, a benchmark designed to test AI agent…
-
DGM and PINN Algorithms Proven to Converge to PDE Solutions
Researchers have mathematically proven that the Deep Galerkin Method (DGM) and Physics Informed Neural Networks (PINNs) can reliably converge to the correct solution for a specific class of semi-linear partial different…
-
New Fourier Neural Operator Extension Tackles Complex PDEs
Researchers have developed an extension to Fourier Neural Operators (FNOs) designed to better model parameterized and coupled partial differential equations (PDEs). The proposed methods incorporate a hypernetwork-based …
-
New physics-informed AI frameworks tackle PDEs with faster training and geometric generalization
Researchers have developed two novel frameworks for physics-informed machine learning to solve partial differential equations (PDEs). The first, Physics-Informed Broad Learning System (PI-BLS), utilizes a backpropagatio…
-
New hierarchical physics-embedded learning approach reduces extrapolation errors by 70%
Researchers have developed a novel hierarchical physics-embedded learning approach that leverages partially known physical laws for spatiotemporal systems. This method encodes known physical structures and their governi…
-
New SDZE framework enables training of 10M-dimensional PINNs on single GPU
Researchers have developed a new framework called the Stochastic Dimension-free Zeroth-order Estimator (SDZE) to address memory and computational constraints in training physics-informed neural networks (PINNs). SDZE ac…
-
FlashPDE library accelerates neural PDE solvers with fused Triton operators
Researchers have developed FlashPDE, a new library of fused Triton operators designed to accelerate the training of physics-informed neural networks (PINNs) for solving partial differential equations (PDEs). This librar…
-
Adaptive Mamba Neural Operators Offer New Paradigm for Solving PDEs
Researchers have introduced Adaptive Mamba Neural Operators (AMO), a novel framework for solving partial differential equations (PDEs) across various geometries and meshes. AMO integrates reproducing kernels with state-…
-
New research tackles PINN training failures with learned initialization and LLM-guided design
Two new research papers explore methods to improve the training and design of physics-informed neural networks (PINNs). The first paper introduces LIGO-PINN, a framework that uses learned initialization to overcome conv…
-
New hybrid method enhances neural operators for complex multiscale problems
Researchers have developed LOD-MSNO, a novel hybrid approach that combines the LOD method with neural operators to address challenges in solving multiscale problems. This method aims to improve the accuracy of neural op…
-
New framework enhances physics-informed neural networks with information propagation paths
Researchers have developed a new framework for physics-informed neural networks (PINNs) that addresses limitations in how these networks handle partial differential equations. The proposed multi-dimensional training-pri…