Table 10 Performance comparison of the predictive solution and predictive parameter of Eq. (44).

From: Data-driven solutions and parameter estimations of a family of higher-order KdV equations based on physics informed neural networks

Case

PINN(\(\tanh\))

PINN(\(\sin\)e)

Loss

9.7835 × 10−7

2.7736 × 10−7

\(\mathbb {L}_{2}\) Error u

2.8395 × 10−4

9.6103 × 10−5

Error \(\alpha\)

0.0338%

0.0371%

\(\alpha\)

1.00034

1.00037

Back to article page