Table 13 Performance comparison of the predictive solution and predictive parameter of Eq. (53).

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

2.3444 × 10−7

1.1147 × 10−6

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

6.8690 × 10−4

1.8833 × 10−3

Error \(\alpha\)

0.0025%

0.2154%

\(\alpha\)

0.99997

1.00215

Back to article page