PINN gradient conflicts occur in distinct regimes (persistent directional, magnitude imbalance, or low/transient) that each favor different fixes, with per-loss adapters plus reweighting improving results on forward and multi-physics problems.
Ali Heydari, Craig A
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 1
baseline 1
citation-polarity summary
verdicts
UNVERDICTED 8representative citing papers
Loss reweighting is cast as an inverse problem that dynamically infers class weights to equalize per-class average losses under the Neural Collapse simplex ETF target.
G-PINNs adds a Gaussian spatial weighting to the PINN loss that detects and tracks shocks from the residual alone, cutting L2 error from 45% to 13% on stationary and 33% to 14% on moving low-viscosity Burgers shocks.
PINNs with hard and soft boundary enforcement solve membrane form-finding PDEs to accuracy comparable with FEM, with hard-BC yielding smaller boundary errors.
An auto-adaptive sampling technique for PINNs is introduced and tested on Allen-Cahn equations to better resolve interfacial regions compared to residual-adaptive methods.
Reviews linear and nonlinear SciML surrogates for coupled fluid flow and transport, with new PINN modeling of turbidity currents and β-VAE mode extraction from Rayleigh-Bénard convection.
PINN achieves 91% accuracy in 3D noisy heat diffusion vs 36% for FDM and 3.3x better error reduction in physical experiment, with efficiency gains in high dimensions.
A review assessing PINN advances for forward modeling, inverse design, and equation discovery across multi-physics domains.
citing papers explorer
-
Per-Loss Adapters for Gradient Conflict in Physics-Informed Neural Networks
PINN gradient conflicts occur in distinct regimes (persistent directional, magnitude imbalance, or low/transient) that each favor different fixes, with per-loss adapters plus reweighting improving results on forward and multi-physics problems.
-
Rethinking Loss Reweighting for Imbalance Learning as an Inverse Problem: A Neural Collapse Point of View
Loss reweighting is cast as an inverse problem that dynamically infers class weights to equalize per-class average losses under the Neural Collapse simplex ETF target.
-
G-PINNs: Gaussian-based spatially weighted formulation for PINNs: 1D low-viscous Burgers
G-PINNs adds a Gaussian spatial weighting to the PINN loss that detects and tracks shocks from the residual alone, cutting L2 error from 45% to 13% on stationary and 33% to 14% on moving low-viscosity Burgers shocks.
-
Physics-informed neural networks for form-finding of unilateral membrane structures
PINNs with hard and soft boundary enforcement solve membrane form-finding PDEs to accuracy comparable with FEM, with hard-BC yielding smaller boundary errors.
-
Auto-Adaptive PINNs with Applications to Phase Transitions
An auto-adaptive sampling technique for PINNs is introduced and tested on Allen-Cahn equations to better resolve interfacial regions compared to residual-adaptive methods.
-
Advances in Scientific Machine Learning for Coupled Fluid Flow and Transport
Reviews linear and nonlinear SciML surrogates for coupled fluid flow and transport, with new PINN modeling of turbidity currents and β-VAE mode extraction from Rayleigh-Bénard convection.
-
Overcoming the Limits of Finite Difference Method; Physics-Informed Neural Network for Noisy High-Dimensional Heat Diffusion
PINN achieves 91% accuracy in 3D noisy heat diffusion vs 36% for FDM and 3.3x better error reduction in physical experiment, with efficiency gains in high dimensions.
-
Beyond Data-Driven: How Physics-Informed Neural Networks are Reshaping Multi-Physics Design and Discovery
A review assessing PINN advances for forward modeling, inverse design, and equation discovery across multi-physics domains.