English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Recently, optical neural networks (ONNs) integrated in photonic chips hasreceived extensive attention because they are expected to implement the samepattern recognition tasks in the electronic platforms with high efficiency andlow power consumption. However, the current lack of various learning algorithmsto train the ONNs obstructs their further development. In this article, wepropose a novel learning strategy based on neuroevolution to design and trainthe ONNs. Two typical neuroevolution algorithms are used to determine thehyper-parameters of the ONNs and to optimize the weights (phase shifters) inthe connections. In order to demonstrate the effectiveness of the trainingalgorithms, the trained ONNs are applied in the classification tasks for irisplants dataset, wine recognition dataset and modulation formats recognition.The calculated results exhibit that the training algorithms based onneuroevolution are competitive with other traditional learning algorithms onboth accuracy and stability. Compared with previous works, we introduce anefficient training method for the ONNs and demonstrate their broad applicationprospects in pattern recognition, reinforcement learning and so on.

Efficient training and design of photonic neural network through neuroevolution