Enhanced Image Recognition Using Gaussian Boson Sampling

TL;DR

This paper demonstrates the application of Gaussian boson sampling (GBS) with a large-scale photonic processor to improve image recognition accuracy on MNIST datasets, showcasing its potential for practical machine learning tasks.

Contribution

The work introduces a GBS-based image recognition scheme inspired by extreme learning machines, implemented on the Jiuzhang device, achieving high accuracy and surpassing classical methods.

Findings

Achieved 95.86% accuracy on MNIST

Surpassed classical SVC with linear kernel

Demonstrated GBS's potential in real-world ML applications

Abstract

Gaussian boson sampling (GBS) has emerged as a promising quantum computing paradigm, demonstrating its potential in various applications. However, most existing works focus on theoretical aspects or simple tasks, with limited exploration of its capabilities in solving real-world practical problems. In this work, we propose a novel GBS-based image recognition scheme inspired by extreme learning machine (ELM) to enhance the performance of perceptron and implement it using our latest GBS device, Jiuzhang. Our approach utilizes an 8176-mode temporal-spatial hybrid encoding photonic processor, achieving approximately 2200 average photon clicks in the quantum computational advantage regime. We apply this scheme to classify images from the MNIST and Fashion-MNIST datasets, achieving a testing accuracy of 95.86% on MNIST and 85.95% on Fashion-MNIST. These results surpass those of classical method SVC with linear kernel and previous physical ELM-based experiments. Additionally, we explore the influence of three hyperparameters and the efficiency of GBS in our experiments. This work not only demonstrates the potential of GBS in real-world machine learning applications but also aims to inspire further advancements in powerful machine learning schemes utilizing GBS technology.