Active Physics Informed Deep Learning: Surrogate Modeling for Non Planar Wavefront Excitation of Topological Nanophotonic Devices
arXiv:2410.16905 · doi:10.1021/acs.nanolett.4c05120
Abstract
Topological plasmonics offers new ways to manipulate light by combining concepts from topology and plasmonics, similar to topological edge states in photonics. However, designing such topological states remains challenging due to the complexity of the high dimensional design space. We present a novel method that uses supervised, physics informed deep learning and surrogate modeling to design topological devices for specific wavelengths. By embedding physical constrains in the neural network training, our model efficiently explores the design space, significantly reducing simulation time. Additionally, we use non planar wavefront excitations via electron beams to probe topologically protected plasmonic modes, making the design and training process nonlinear. Using this approach, we design a topological device with unidirectional edge modes in a ring resonator at specific operational frequencies. Our method reduces computational cost and time while maintaining high accuracy, highlighting the potential of combining machine learning and advanced techniques for photonic device innovation.
References in corpus (19)
- Topological Photonics
- Training deep neural networks for the inverse design of nanophotonic structures
- Plasmons in nearly touching metallic nanoparticles: singular response in the limit of touching dimers
- Engineering of robust topological quantum phases in graphene nanoribbons
- Observation of the topological soliton state in the Su-Schrieffer-Heeger model
- Deep learning in nano-photonics: inverse design and beyond
- Observation of the spin-based plasmonic effect in nanoscale structures
- Observation of Dirac plasmons in a topological insulator
- Topological Edge Plasmon Modes between Diatomic Chains of Nanoparticles
- The effects of retardation on the topological plasmonic chain: plasmonic edge states beyond the quasistatic limit
- Physics-informed linear regression is competitive with two Machine Learning methods in residential building MPC
- Active learning of deep surrogates for PDEs: Application to metasurface design
- Coherent topological polariton laser
- A topological encoding method for data-driven photonics inverse design
- Green function surface-integral method for nonlocal response of plasmonic nanowires in arbitrary dielectric environments
- Topological plasmons in dimerized chains of nanoparticles: robustness against long-range quasistatic interactions and retardation effects
- Deep learning enabled strategies for modelling of complex aperiodic plasmonic metasurfaces of arbitrary size
- Plasmon-exciton Interactions in Gold-WSe2 Multilayer structures
- Topological photonics by breaking the degeneracy of line node singularities in semimetal-like photonic crystals