Efficient training and design of photonic neural network through neuroevolution

TL;DR

This paper introduces a neuroevolution-based training strategy for photonic neural networks, enabling efficient hyper-parameter tuning and weight optimization, demonstrated on various classification tasks with competitive accuracy and stability.

Contribution

It presents a novel neuroevolution approach for training ONNs, addressing the lack of effective learning algorithms and expanding their application potential.

Findings

Neuroevolution algorithms achieve high accuracy in classification tasks.

The proposed method is competitive with traditional training algorithms.

Demonstrated broad application prospects in pattern recognition and reinforcement learning.

Abstract

Recently, optical neural networks (ONNs) integrated in photonic chips has received extensive attention because they are expected to implement the same pattern recognition tasks in the electronic platforms with high efficiency and low power consumption. However, the current lack of various learning algorithms to train the ONNs obstructs their further development. In this article, we propose a novel learning strategy based on neuroevolution to design and train the ONNs. Two typical neuroevolution algorithms are used to determine the hyper-parameters of the ONNs and to optimize the weights (phase shifters) in the connections. In order to demonstrate the effectiveness of the training algorithms, the trained ONNs are applied in the classification tasks for iris plants dataset, wine recognition dataset and modulation formats recognition. The calculated results exhibit that the training algorithms based on neuroevolution are competitive with other traditional learning algorithms on both accuracy and stability. Compared with previous works, we introduce an efficient training method for the ONNs and demonstrate their broad application prospects in pattern recognition, reinforcement learning and so on.