Enhanced Image Recognition Using Gaussian Boson Sampling

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Gong, Si-Qiu, Chen, Ming-Cheng, Liu, Hua-Liang, Su, Hao, Gu, Yi-Chao, Tang, Hao-Yang, Jia, Meng-Hao, Deng, Yu-Hao, Wei, Qian, Wang, Hui, Zhong, Han-Sen, Jiang, Xiao, Li, Li, Liu, Nai-Le, Lu, Chao-Yang, Pan, Jian-Wei
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911020537085952
author Gong, Si-Qiu
Chen, Ming-Cheng
Liu, Hua-Liang
Su, Hao
Gu, Yi-Chao
Tang, Hao-Yang
Jia, Meng-Hao
Deng, Yu-Hao
Wei, Qian
Wang, Hui
Zhong, Han-Sen
Jiang, Xiao
Li, Li
Liu, Nai-Le
Lu, Chao-Yang
Pan, Jian-Wei
author_facet Gong, Si-Qiu
Chen, Ming-Cheng
Liu, Hua-Liang
Su, Hao
Gu, Yi-Chao
Tang, Hao-Yang
Jia, Meng-Hao
Deng, Yu-Hao
Wei, Qian
Wang, Hui
Zhong, Han-Sen
Jiang, Xiao
Li, Li
Liu, Nai-Le
Lu, Chao-Yang
Pan, Jian-Wei
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2506_19707
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhanced Image Recognition Using Gaussian Boson Sampling
Gong, Si-Qiu
Chen, Ming-Cheng
Liu, Hua-Liang
Su, Hao
Gu, Yi-Chao
Tang, Hao-Yang
Jia, Meng-Hao
Deng, Yu-Hao
Wei, Qian
Wang, Hui
Zhong, Han-Sen
Jiang, Xiao
Li, Li
Liu, Nai-Le
Lu, Chao-Yang
Pan, Jian-Wei
Quantum Physics
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.
title Enhanced Image Recognition Using Gaussian Boson Sampling
topic Quantum Physics
url https://arxiv.org/abs/2506.19707