CGMPINN combines Gaussian mixture modeling with curriculum learning to reduce training errors in physics-informed neural networks by up to 97.8% on benchmark PDEs while providing theoretical convergence guarantees.
A stacked adaptive residual PINN (STAR- PINN) approach to 2D time-domain magnetic diffu- sion in nonlinear materials.IEEE Access, 13:141380– 141394, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From Simple to Complex: Curriculum-Guided Physics-Informed Neural Networks via Gaussian Mixture Models
CGMPINN combines Gaussian mixture modeling with curriculum learning to reduce training errors in physics-informed neural networks by up to 97.8% on benchmark PDEs while providing theoretical convergence guarantees.