Reducing Computation Complexity in Optimization of Nanophotonic Structures Using Pruning

Inverse design of complex nanophotonic devices is a very computation-consuming task. Deep-learning-based approaches can facilitate this process. However, due to the lack of solid knowledge about the underlying complexity of the input-output relation for a selected class of nanostructures, it is comm...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE photonics technology letters Jg. 36; H. 4; S. 215 - 218
Hauptverfasser: Hadigheh Javani, Mohammad, Zandehshahvar, Mohammadreza, Adibi, Ali
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York IEEE 15.02.2024
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Schlagworte:
Artificial neural networks
Complexity
Computation
computation complexity
Computational complexity
Deep learning
Inverse design
Machine learning
Metasurfaces
Modelling
Nanophotonics
Nanostructure
Nanostructures
Neural networks
Pruning
Weak nodes
ISSN:1041-1135, 1941-0174
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Inverse design of complex nanophotonic devices is a very computation-consuming task. Deep-learning-based approaches can facilitate this process. However, due to the lack of solid knowledge about the underlying complexity of the input-output relation for a selected class of nanostructures, it is common to select an over-parameterized neural network (NN) for modeling this relation. We present a novel pruning method based on removing weak nodes and connections in the original NN to simplify the input-output relation without imposing significant error. In addition to reducing the model complexity computations, the pruned network can be used to find valuable insight into the physics of device operation. To show the efficacy of our approach, we use it for modeling and inverse design of two classes of nanostructures with different complexities.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1041-1135
1941-0174
DOI:10.1109/LPT.2023.3342631