Phase space topology of four-wave mixing reconstructed by a neural network

Article Optics

Idealized four-wave mixing dynamics in a nonlinear Schrodinger equation fiber system

Anastasiia Sheveleva et al.

Summary: In this study, we overcome the challenges of observing ideal four-wave mixing dynamics and extend the effective propagation distance using programmable phase and amplitude shaping. The experimental results are in excellent agreement with numerical solutions, revealing the complete phase space topology.

OPTICA (2022)

Add to Collection

Article Optics

Physics-Informed Neural Network for Nonlinear Dynamics in Fiber Optics

Xiaotian Jiang et al.

Summary: This study investigates a physics-informed neural network (PINN) that combines deep learning with physics to solve the nonlinear Schrodinger equation in fiber optics. PINN is systematically investigated and verified for multiple physical effects in optical fibers, and it exhibits better performance than data-driven neural networks while using less data. The results show that PINN is not only an effective partial differential equation solver, but also a prospective technique for scientific computing and automatic modeling in fiber optics.

LASER & PHOTONICS REVIEWS (2022)

Add to Collection

Article Multidisciplinary Sciences

Data-driven model discovery of ideal four-wave mixing in nonlinear fibre optics

Andrei Ermolaev et al.

Summary: Using sparse regression, we successfully recovered the governing differential equation model of ideal four-wave mixing in a nonlinear Schrodinger equation optical fiber system from dynamical data. Analysis of ensemble data allowed us to reliably identify the governing model in the presence of noise.

SCIENTIFIC REPORTS (2022)

Add to Collection

Review Optics

Machine learning and applications in ultrafast photonics

Goery Genty et al.

Summary: The field of smart photonics has seen rapid growth and development in recent years, with machine-learning algorithms being used to enhance optical systems and add new functionalities. Machine learning in ultrafast photonics has shown potential in accelerating technology, particularly in areas such as pulsed laser design, operation, and ultrafast propagation dynamics. However, challenges and future research areas in this field still need to be addressed.

NATURE PHOTONICS (2021)

Add to Collection

Article Multidisciplinary Sciences