What question did this study set out to answer?

March 21, 2026

Monte Carlo Optimization for Real-Time Magnetic Domain Learning in Magneto-Optical Diffractive Deep Neural Networks

Key Points

The research aims to optimize learning in magneto-optical diffractive deep neural networks using a Monte Carlo-based method.
Developed a Monte Carlo method tailored for online learning in MO-D2NN.
Minimized cross-entropy loss without gradients at a lower computational cost.
Implemented the algorithm in a physical optical setup for validation.
Achieved 96% accuracy in classifying MNIST handwritten digits with a single hidden layer.
Demonstrated robust performance as an alternative to traditional gradient descent methods.
Validated real-time learning capabilities in an experimental optical setup.

Abstract

We propose an optimized algorithm using Monte Carlo Method (MCM) tailored for online learning in magneto-optical diffractive deep neural networks (MO-D2NN), a physical neural network platform where binary-weight are determined by magneto-optical modulation of light through the Faraday effect. Our derivative-free approach based MCM, iteratively adjusts the magnetic domain patterns to minimize cross-entropy loss without relying on gradients at a much lower computational cost. Our findings reveal that the MCM-based optimization algorithm serves as a robust and viable alternative to gradient descent-based training, achieving an accuracy of 96% for MNIST handwritten digits classification with only a single hidden layer, highlighting its potential as a powerful approach for training MO-D2NN. We further validate its feasibility through physical implementation in an experimental optical setup, confirming its practical applicability for online image recognition tasks. We successfully demonstrate real-time learning of MO-D2NN using the MCM algorithm.

AI에게 질문

Bookmark