An adaptive KAN-based PINN framework for axisymmetric pulsar magnetosphere achieves O(1e-6) PDE residual errors, under-20-minute convergence, smaller stellar radii, and a correction to the flux-T-point equation.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
verdicts
UNVERDICTED 5representative citing papers
Spectral deflation anchored to a single reference Schur complement reduces CG iterations 55-98% across diffusion, convection-diffusion, and heat-transfer benchmarks by restricting low eigenmodes to varying inactive sets.
DiffUNet^2 is a bidirectional conditional diffusion model integrated with visual tools for probabilistic exploration of scientific time series across five evaluated datasets.
ChainzRule with DREG regularization claims 15.5x fewer parameters than standard models, 23.1% lower peak gradient volatility on MNIST, and 70.17% accuracy on Yelp Full ordinal regression.
Neural network reconstructs cosmic-ray trajectories to better than 1.4° angular resolution and separates charges to >95% accuracy for Z≤8 using Geant4-simulated data for the RadMap Telescope.
citing papers explorer
-
An adaptive framework for the axisymmetric pulsar magnetosphere using physics-informed Kolmogorov-Arnold networks
An adaptive KAN-based PINN framework for axisymmetric pulsar magnetosphere achieves O(1e-6) PDE residual errors, under-20-minute convergence, smaller stellar radii, and a correction to the flux-T-point equation.
-
Online Spectral Deflation for State Constrained Optimal Control Problems
Spectral deflation anchored to a single reference Schur complement reduces CG iterations 55-98% across diffusion, convection-diffusion, and heat-transfer benchmarks by restricting low eigenmodes to varying inactive sets.
-
DiffUNet^2: Bidirectional Prediction, Probabilistic Generation and Collaborative Visual Discovery for Scientific Data
DiffUNet^2 is a bidirectional conditional diffusion model integrated with visual tools for probabilistic exploration of scientific time series across five evaluated datasets.
-
Layer-wise Derivative Controlled Networks
ChainzRule with DREG regularization claims 15.5x fewer parameters than standard models, 23.1% lower peak gradient volatility on MNIST, and 70.17% accuracy on Yelp Full ordinal regression.
-
A Neural-Network Framework for Tracking and Identification of Cosmic-Ray Nuclei in the RadMap Telescope
Neural network reconstructs cosmic-ray trajectories to better than 1.4° angular resolution and separates charges to >95% accuracy for Z≤8 using Geant4-simulated data for the RadMap Telescope.