유전 알고리즘 기반 양자 자동인코더 아키텍처 탐색
📝 Abstract
In recent years, machine learning and deep learning have driven advances in domains such as image classification, speech recognition, and anomaly detection by leveraging multi-layer neural networks to model complex data. Simultaneously, quantum computing (QC) promises to address classically intractable problems via quantum parallelism, motivating research in quantum machine learning (QML). Among QML techniques, quantum autoencoders show promise for compressing high-dimensional quantum and classical data. However, designing effective quantum circuit architectures for quantum autoencoders remains challenging due to the complexity of selecting gates, arranging circuit layers, and tuning parameters. This paper proposes a neural architecture search (NAS) framework that automates the design of quantum autoencoders using a genetic algorithm (GA). By systematically evolving variational quantum circuit (VQC) configurations, our method seeks to identify high-performing hybrid quantum-classical autoencoders for data reconstruction without becoming trapped in local minima. We demonstrate effectiveness on image datasets, highlighting the potential of quantum autoencoders for efficient feature extraction within a noise-prone, near-term quantum era. Our approach lays a foundation for broader application of genetic algorithms to quantum architecture search, aiming for a robust, automated method that can adapt to varied data and hardware constraints.
💡 Analysis
In recent years, machine learning and deep learning have driven advances in domains such as image classification, speech recognition, and anomaly detection by leveraging multi-layer neural networks to model complex data. Simultaneously, quantum computing (QC) promises to address classically intractable problems via quantum parallelism, motivating research in quantum machine learning (QML). Among QML techniques, quantum autoencoders show promise for compressing high-dimensional quantum and classical data. However, designing effective quantum circuit architectures for quantum autoencoders remains challenging due to the complexity of selecting gates, arranging circuit layers, and tuning parameters. This paper proposes a neural architecture search (NAS) framework that automates the design of quantum autoencoders using a genetic algorithm (GA). By systematically evolving variational quantum circuit (VQC) configurations, our method seeks to identify high-performing hybrid quantum-classical autoencoders for data reconstruction without becoming trapped in local minima. We demonstrate effectiveness on image datasets, highlighting the potential of quantum autoencoders for efficient feature extraction within a noise-prone, near-term quantum era. Our approach lays a foundation for broader application of genetic algorithms to quantum architecture search, aiming for a robust, automated method that can adapt to varied data and hardware constraints.
📄 Content
Neural Architecture Search for Quantum Autoencoders Hibah Agha∗¶, Samuel Yen-Chi Chen †∥, Huan-Hsin Tseng‡∗∗, Shinjae Yoo‡†† ∗College of Engineering and Computing Sciences, New York Institute of Technology, Old Westbury, NY 11568, USA †Wells Fargo, New York, NY 10017, USA ‡AI Department, Brookhaven National Laboratory, Upton NY, USA Email:¶ hibahswe@gmail.com, ∥yen-chi.chen@wellsfargo.com, ∗∗htseng@bnl.gov, †† sjyoo@bnl.gov Abstract—In recent years, machine learning and deep learning have driven advances in domains such as image classification, speech recognition, and anomaly detection by leveraging multi- layer neural networks to model complex data. Simultaneously, quantum computing (QC) promises to address classically in- tractable problems via quantum parallelism, motivating research in quantum machine learning (QML). Among QML techniques, quantum autoencoders show promise for compressing high- dimensional quantum and classical data. However, designing effective quantum circuit architectures for quantum autoencoders remains challenging due to the complexity of selecting gates, arranging circuit layers, and tuning parameters. This paper proposes a neural architecture search (NAS) frame- work that automates the design of quantum autoencoders using a genetic algorithm (GA). By systematically evolving variational quantum circuit (VQC) configurations, our method seeks to iden- tify high-performing hybrid quantum-classical autoencoders for data reconstruction without becoming trapped in local minima. We demonstrate effectiveness on image datasets, highlighting the potential of quantum autoencoders for efficient feature extraction within a noise-prone, near-term quantum era. Our approach lays a foundation for broader application of genetic algorithms to quantum architecture search, aiming for a robust, automated method that can adapt to varied data and hardware constraints. Index Terms—Quantum Machine Learning, Quantum Neural Networks, Variational Quantum Circuits, Autoencoder, Quantum Architecture Search I. INTRODUCTION Deep learning (DL) has become foundational to a wide range of modern applications, including computer vision, nat- ural language processing, and scientific research. Its strength lies in its ability to uncover complex patterns in data and produce highly accurate predictions [1]. DL builds upon traditional machine learning (ML) by employing deep neural network architectures—comprised of multiple interconnected layers, to extract features and learn representations at various levels of abstraction. These capabilities have driven remarkable progress in tasks once deemed infeasible [2] such as mastering complex games like Go through reinforcement learning de- The views expressed in this article are those of the authors and do not represent the views of Wells Fargo. This article is for informational purposes only. Nothing contained in this article should be construed as investment advice. Wells Fargo makes no express or implied warranties and expressly disclaims all legal, tax, and accounting implications related to this article. Fig. 1. Hybrid Quantum-Classical Computing Paradigm. feating world champions [3], and predicting protein structures with atomic accuracy [4]. Concurrently, quantum computing (QC) is rapidly evolving from theory to practice, with companies like Google, IBM, and Intel developing hardware to achieve quantum advantage [5]. QC offers the potential to solve problems that are intractable for classical computers by leveraging quantum superposition and entanglement, particularly in optimization, machine learn- ing, and simulation tasks [6], [7]. Despite these promising capabilities, current quantum de- vices—referred to as Noisy Intermediate-Scale Quantum (NISQ) devices—face significant challenges. These include susceptibility to noise, decoherence, and restricted qubit counts, necessitating error correction techniques [8]–[10]. To navigate these limitations, hybrid quantum-classical comput- ing has emerged as a practical and scalable approach. One of the most prominent strategies for tackling such limitations are variational quantum algorithms (VQAs) [11], by leveraging quantum resources where beneficial while relying on classical systems for robustness. The general idea of the hybrid com- puting paradigm is illustrated in Figure 1. Building on the foundation of VQAs, the intersection of QC and ML, quantum machine learning (QML), has garnered sig- nificant research interest due to its potential to outperform clas- sical machine learning models [12]–[14]. Empirically, QML has demonstrated successful use cases in various domains such as classification [15]–[21], sequence learning [22]–[24], natu- ral language processing [25]–[28] and reinforcement learning [29]–[34]. However, certain specialized applications, such as arXiv:2511.19246v1 [quant-ph] 24 Nov 2025 quantum autoencoders, deliver a promising, yet relatively underexplored application. By leveraging quantum principles, instead
This content is AI-processed based on ArXiv data.