Theory and Modern Applications
Table 5 IVP in Eq. (29). Results for a complete learning procedure for one subdomain
From: Adaptive neural-domain refinement for solving time-dependent differential equations
\(t_{25}\) | \(t_{26}\) | \(\Delta \psi _{1}\) | \(E_{25}^{\mathrm{TP}}\) | \(E_{25}^{\mathrm{VP}}\) | α |
|---|---|---|---|---|---|
5.5953 | 15.000 | 1.3557 | 22.888 | 23.598 | 1e-3 |
5.5953 | 10.298 | 2.3780 | 1.7622 | 22.695 | 1e-3 |
5.5953 | 7.9465 | 1.3774 | 6.8529 | 34.448 | 1e-3 |
5.5953 | 7.9465 | 0.3868 | 10.640 | 9.9730 | 6e-3 |
5.5953 | 6.7709 | 3.4447e-2 | 3.2209e-2 | 3.9242e-2 | 6e-3 |
5.5953 | 6.1831 | 1.1881e-2 | 3.4791e-2 | 3.5682e-2 | 6e-3 |
5.5953 | 5.8892 | 3.3488e-4 | 1.0875e-4 | 1.0779e-4 | 6e-3 |
5.5953 | 5.7423 | 1.6882e-4 | 2.4565e-6 | 2.3747e-6 | 6e-3 |