Solving 1D Poisson Equation (alpha1=5.0, alpha2=3.0, s=20.0) (test)
0.001eLossCGMPINN
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| CGMPINNOptimizer=Adam→L-BFGS, Network Architecture=4-hidden-layer network with 50 neurons per layer, Training Points=1,500 randomly sampled interior collocation points, Evaluation Points=200 uniformly distributed test points2026.05 | 0.001 | 0.0002 | 0.0002 | 0.0004 | 895.9 | |
| PINNOptimizer=Adam→L-BFGS, Network Architecture=4-hidden-layer network with 50 neurons per layer, Training Points=1,500 randomly sampled interior collocation points, Evaluation Points=200 uniformly distributed test points2026.05 | 0.0019 | 0.0088 | 0.0082 | 0.0092 | 1,018.2 | |
| LNN-PINNOptimizer=Adam→L-BFGS, Network Architecture=4-hidden-layer network with 50 neurons per layer, Training Points=1,500 randomly sampled interior collocation points, Evaluation Points=200 uniformly distributed test points2026.05 | 0.0023 | 0.0005 | 0.0004 | 0.0008 | 1,422 | |
| STAR-PINNOptimizer=Adam→L-BFGS, Network Architecture=4-hidden-layer network with 50 neurons per layer, Training Points=1,500 randomly sampled interior collocation points, Evaluation Points=200 uniformly distributed test points2026.05 | 0.0687 | 0.0012 | 0.0011 | 0.0021 | 1,332.4 | |
| gPINNOptimizer=Adam→L-BFGS, Network Architecture=4-hidden-layer network with 50 neurons per layer, Training Points=1,500 randomly sampled interior collocation points, Evaluation Points=200 uniformly distributed test points2026.05 | 0.102 | 0.0188 | 0.0174 | 0.0247 | 5,711.8 | |
| lbPINNOptimizer=Adam→L-BFGS, Network Architecture=4-hidden-layer network with 50 neurons per layer, Training Points=1,500 randomly sampled interior collocation points, Evaluation Points=200 uniformly distributed test points2026.05 | 2.95 | 4.57 | 4.22 | 7.88 | 1,205.6 |