Table 12 Performance comparison of the predictive solution and predictive parameter of Eq. (50).

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

7.3111 × 10−7

9.7328 × 10−7

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

5.7212 × 10−3

5.07647 × 10−3

Error \(\alpha\)

8.1902%

3.7163%

\(\alpha\)

29.9181

30.037

Back to article page