CSD-PINN: a complex-symmetric, dynamic-weighted physics-informed neural network for parameter identification in high-order nonlinear Schrödinger equations

Key Points

• Effective parameter identification achieved through a novel physics-informed neural network.
• The approach accurately solves high-order nonlinear Schrödinger equations, indicating a promising new method.
• This method employs dynamic weighting and complex symmetry to enhance identification precision.
• Highlights potential applications in various fields, such as quantum mechanics and photonics, supporting further advancements.