Modeling and benchmarking quantum optical neurons for efficient neural computation

2025-09-01

AI Summary

This study addresses the feasibility of quantum optical neurons (QONs) for efficient neural computation. We propose a differentiable photonic neuron architecture leveraging Hong–Ou–Mandel and Mach–Zehnder interferometers, unifying phase, amplitude, and intensity modulation within a physics-informed differentiable framework. Our modeling reveals that specific interferometric configurations significantly enhance training convergence speed and stability. Evaluated on MNIST and FashionMNIST image classification tasks, the two top-performing configurations achieve accuracy comparable to—or exceeding—that of classical fully connected neurons; in contrast, intensity-encoded variants, while promising, exhibit training instability. These findings establish a novel design paradigm for photonic neural networks and advance the practical realization of QONs as core components in hybrid optoelectronic intelligent hardware.

Technology Category

• Machine Learning: Quantum Machine Learning
• Search and Optimization: Learning to Search
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Quantum optical neurons (QONs) are emerging as promising computational units that leverage photonic interference to perform neural operations in an energy-efficient and physically grounded manner. Building on recent theoretical proposals, we introduce a family of QON architectures based on Hong-Ou-Mandel (HOM) and Mach-Zehnder (MZ) interferometers, incorporating different photon modulation strategies -- phase, amplitude, and intensity. These physical setups yield distinct pre-activation functions, which we implement as fully differentiable modules in software. We evaluate these QONs both in isolation and as building blocks of multilayer networks, training them on binary and multiclass image classification tasks using the MNIST and FashionMNIST datasets. Our experiments show that two configurations -- HOM-based amplitude modulation and MZ-based phase-shifted modulation -- achieve performance comparable to that of classical neurons in several settings, and in some cases exhibit faster or more stable convergence. In contrast, intensity-based encodings display greater sensitivity to distributional shifts and training instabilities. These results highlight the potential of QONs as efficient and scalable components for future quantum-inspired neural architectures and hybrid photonic-electronic systems.

Problem

• Modeling quantum optical neurons for neural computation
• Benchmarking QON performance on image classification tasks
• Evaluating photon modulation strategies in interferometer architectures

Innovation

• Quantum optical neurons using photonic interference
• Differentiable modules with HOM and MZ interferometers
• Phase and amplitude modulation achieving classical performance