A cross-section-based scaling of the loss function accelerates convergence and improves accuracy for MF-PINNs on neutron diffusion problems across 1D-3D and fixed-source to eigenvalue cases.
When and why PINNs fail to train: A neural tangent kernel perspective.Journal of Computational Physics, 449:110768
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3representative citing papers
PINNs fail on spurious solutions admitted by the residual loss; adaptive pseudo-time stepping with Jacobian-based step selection improves accuracy and robustness on PDE benchmarks.
PINNACLE is an open-source framework for classical and quantum PINNs that supplies modular training methods and benchmarks showing high sensitivity to architecture choices plus parameter-efficiency gains in some hybrid quantum regimes.
citing papers explorer
-
On Physics-Based Loss Scaling for MF-PINNs applied to the neutron diffusion equation
A cross-section-based scaling of the loss function accelerates convergence and improves accuracy for MF-PINNs on neutron diffusion problems across 1D-3D and fixed-source to eigenvalue cases.
-
When PINNs Go Wrong: Pseudo-Time Stepping Against Spurious Solutions
PINNs fail on spurious solutions admitted by the residual loss; adaptive pseudo-time stepping with Jacobian-based step selection improves accuracy and robustness on PDE benchmarks.
-
PINNACLE: An Open-Source Computational Framework for Classical and Quantum PINNs
PINNACLE is an open-source framework for classical and quantum PINNs that supplies modular training methods and benchmarks showing high sensitivity to architecture choices plus parameter-efficiency gains in some hybrid quantum regimes.