IEEE International Midwest Symposium on Circuits and Systems (MWSCAS)

Investigating a Quantum Cloud Paradigm with Quantum Neural Networks

Maxwell Yarter, Glen S. Uehara, Andreas Spanias

Published Aug 6, 2023• 4 citations

Abstract

In this work, we examine the interactions between an embedded processing board and cloud server performing quantum computing simulations. We examined the trade-offs in performance and complexity between using classical neural networks and quantum-hybrid neural networks on the cloud server. More specifically, we propose the use of quantum hybrid neural network (QNN) for the classification of spoken commands on live audio data acquired at the networks edge. For this application, a compact embedded processor board handles simple operations like data acquisition and pre-processing while a quantum cloud server is used to perform quantum computing simulations. This experimental setup is well suited for quantum computing as it is not feasible to embed a quantum processor at the edge, but practical to interface embedded processor boards to a quantum cloud server for processing massive audio data for a variety of recognition tasks. We found that distributing the task between the embedded processor and quantum sever enabled the application of QNN's to live audio data at the expense of system response time.

DOI: 10.1109/mwscas57524.2023.10406072

Related papers

Exploring Quantum Processors in Decentralized Edge Computing for Improved Data Processing

Harshad Pitkar

The rapid growth of data generated by Internet of Things (IoT) devices necessitates the development of advanced computational frameworks that can efficiently handle real-time data processing. Traditional cloud and edge computing systems, while effective, face significant challenges, including high latency, limited computational resources, and data transfer bottlenecks. To address these issues, this study introduces a hybrid approach that integrates quantum computing into the edge-cloud continuum. The proposed method utilizes Quantum- Neural Networks (QNNs) in conjunction with local edge processing, leveraging quantum autoencoders for feature compression, and incorporating fully decentralized edge computing with quantum processors to enhance data processing capabilities. Additionally, a hybrid model combining QNNs with classical machine learning techniques is implemented at the edge for improved classification accuracy. The proposed system's performance is evaluated through extensive experiments across multiple datasets, comparing the results of quantum-based models with traditional classical and hybrid approaches. The results show a significant improvement in classification accuracy, reduced feature compression time, and enhanced processing efficiency, particularly in resource-constrained environments. The results underscore the potential of quantum machine learning in real-time, large-scale applications, marking an important step toward the practical integration of quantum computing in distributed computing systems.

International Conference on Machine Learning and Autonomous Systems (ICMLAS)Mar 10, 20250 citations

Architectural Vision for Quantum Computing in the Edge-Cloud Continuum

Alireza Furutanpey, Johanna Barzen, Marvin Bechtold +4 more

Quantum processing units (QPUs) are currently exclusively available from cloud vendors. However, with recent advancements, hosting QPUs will soon be possible everywhere. Existing work has yet to draw from research in edge computing to explore systems exploiting mobile QPUs, or how hybrid applications can benefit from distributed heterogeneous resources. Hence, this work presents an architecture for Quantum Computing in the edge-cloud continuum. We discuss the necessity, challenges, and solution approaches for extending existing work on classical edge computing to integrate QPUs. We describe how warm-starting allows defining workflows that exploit the hierarchical resources spread across the continuum. Then, we introduce a distributed inference engine with hybrid classical-quantum neural networks (QNNs) to aid system designers in accommodating applications with complex requirements that incur the highest degree of heterogeneity. We propose solutions focusing on classical layer partitioning and quantum circuit cutting to demonstrate the potential of utilizing classical and quantum computation across the continuum. To evaluate the importance and feasibility of our vision, we provide a proof of concept that exemplifies how extending a classical partition method to integrate quantum circuits can improve the solution quality. Specifically, we implement a split neural network with optional hybrid QNN predictors. Our results show that extending classical methods with QNNs is viable and promising for future work.

arXiv:2305.05238IEEE International Conference on Quantum Software (QSW)Jul 1, 202329 citations

Development of quantum neural networks for complex data classification

Asgari Savvas, Mian Snell Lizarralde, Patrisia Teresa Marsoit

This research explores the development of Quantum Neural Networks (QNNs) as a transformative approach for complex data classification. Utilizing a numerical example, we illustrate the foundational quantum principles of superposition and entanglement within QNNs. The hybrid quantum-classical processing paradigm is introduced, emphasizing the seamless integration of quantum and classical components, acknowledging the challenges of quantum error correction and noise in Noisy Intermediate-Scale Quantum (NISQ) devices. While the example is deliberately simple, it serves as a starting point for understanding the unique advantages and challenges associated with QNNs. Our findings highlight the potential of quantum computation for parallel processing but also underscore the need to address current limitations for practical applications. Future research directions include investigating sophisticated quantum circuits, exploring error mitigation strategies, and assessing QNN performance across diverse datasets. Collaboration between quantum computing and machine learning communities is essential for the advancement of QNNs, and developments in quantum hardware will play a pivotal role in realizing their full potential. This study contributes to the evolving discourse at the intersection of quantum computing and machine learning, providing foundational insights and laying the groundwork for further exploration in this rapidly advancing field.

Journal of Computer Science and Research (JoCoSiR)Nov 30, 20232 citations

A Hybrid Approach to Data Analysis: Performance Issues in Integrated Quantum and Cloud Computing

Dr Rajendirakumar S

Abstract The integration of cloud computing architectures with quantum computational models represents a significant paradigm shift in the field of large-scale data analysis. While quantum processors possess the theoretical capability to solve complex statistical and logistical problems exponentially faster than classical systems, deploying them within conventional cloud environments introduces severe performance bottlenecks. These challenges primarily revolve around network latency, quantum state decoherence, and the inadequacies of classical resource provisioning algorithms. In this paper, we explore the performance issues associated with hybrid quantum-cloud infrastructures and propose a structured decision-making and scheduling framework. By analyzing deployment constraints, cryptographic overheads, and hardware limitations, this study highlights the critical pathways and ethical considerations necessary for standardizing Quantum-as-a-Service (QaaS) in modern enterprise datacenters. Keywords: Quantum processors, Quantum-as-a-Service(QaaS), makespan 1.Introduction The integration of cloud computing and quantum computation represents a crucial frontier for the future of complex data analysis. Cloud computing has revolutionized the modern information technology landscape by transitioning computational resources from localized, purchased products to highly scalable, on-demand services delivered over the internet. Simultaneously, quantum computing offers unprecedented computational advantages through fundamental physical phenomena like entanglement and superposition, enabling the resolution of highly complex logistical, chemical, and statistical problems. However, merging these two distinct technological domains into a unified framework introduces severe performance bottlenecks related to dynamic resource provisioning, network communication latency, and task scheduling. The primary problem lies in the efficient execution of large-scale data analysis pipelines that make both classical datacenters and quantum processing units (QPUs). Existing approaches are demonstrably insufficient for managing this hybrid computing environment for two main reasons. First, conventional cloud load balancing and resource allocation algorithms are designed solely for deterministic classical virtual machines, fundamentally failing to account for the unique coherence times, error rates, and state preparation constraints of quantum hardware. Fig 1.1 shows Cloud balancing. Second,current cryptographic protocols designed for delegated quantum computing , while excellent at ensuring the blindness and security of user data, impose substantial communication and entangling overheads that severely degrade end-to-end performance in real-world datacenter along with Circuit diagram in Fig.1.2. Figure 1.1 Cloud Balancing with web servers To address these critical performance issues, this paper introduces a novel architectural approach designed specifically for hybrid quantum-cloud systems. We aim to bridge the gap between theoretical quantum potential and practical cloud deployment. Specifically, our work makes the following contributions: We propose a structured, hybrid decision-making and scheduling framework that dynamically partitions and routes data analysis tasks between classical cloud servers and quantum nodes. We formulate a comprehensive, hypothetical evaluation methodology to simulate the makespan and resource utilization of this hybrid system, identifying key latency bottlenecks in delegated quantum operations. Fig 1.2. Quantum Circuit diagram 2.Classical Cloud Computing The first category of related work focuses on resource management and provisioning in classical cloud computing. Traditional methods in this domain emphasize minimizing the computational makespan of tasks through selective heuristics, such as min-min and max-min scheduling algorithms, which allocate virtual machines based on fluctuating task demands. Furthermore, comprehensive load balancing schemes have been developed to ensure efficient resource utilization in accordance with strict Service Level Agreements (SLAs), allowing resources to be provisioned on a pay-as-you-go basis. While these classical models demonstrate robust strengths in minimizing computational wait times and maximizing server efficiency, their fundamental weakness is their inability to schedule probabilistic and highly sensitive quantum operations. In contrast to these purely classical frameworks, our proposed work directly integrates quantum-specific hardware constraints, such as state preparation delays and decoherence windows, into the provisioning logic. The second category addresses broader enterprise cloud adoption and organizational decision-making frameworks. Research in this area highlights the necessity for generic, multi-criteria decision frameworks that evaluate cloud infrastructures beyond mere cost considerations, factoring in specific enterprise risks, benefits, and IT support metrics. Additionally, numerous studies have developed strategic roadmaps for Small and Medium Enterprises (SMEs) to navigate complex interoperability, knowledge, and security challenges during cloud migration. Although these frameworks provide excellent strategic guidance for adopting standard classical services, they lack operational mechanisms for managing the performance issues introduced by bleeding-edge quantum co-processors. Our work builds upon these established enterprise decision-making structures by providing a functional, low-level deployment pipeline tailored for hybrid operations. The third category explores the theoretical foundations of delegated quantum computing and interdisciplinary quantum applications. Delegated quantum computing allows clients with extremely low computational capabilities to securely outsource demanding tasks to a powerful quantum server, often utilizing measurement-based models and complex prepare-and-send communication paradigms to preserve data blindness. Concurrently, quantum computing frameworks have been utilized to explore deep interdisciplinary problems—ranging from physics paradoxes to advanced chemical modeling—demonstrating vast analytical potential. The primary weakness of this category is that much of the academic research remains focused on abstract theoretical protocols or isolated quantum circuits, rather than addressing the physical infrastructure pitfalls and massive scale required for cloud integration. This paper bridges that gap by embedding delegated quantum paradigms within a concrete, performance-oriented enterprise cloud architecture. 3.Hybrid Quantum-Cloud To tackle the performance issues inherent in hybrid data analysis, we propose a modular Hybrid Quantum-Cloud Provisioning Framework. The system follows a structured, numbered pipeline designed to seamlessly route computational tasks across heterogeneous hardware. First, classical user requests are ingested by a central cloud load balancer that analyzes the computational complexity and algorithmic requirements of the incoming dataset. Second, the system employs a multi-criteria decision engine to partition the workflow; classical data preprocessing is routed to standard virtual machines, while complex optimization components are packaged as isolated quantum circuits. Third, the framework utilizes a delegated quantum computing protocol to securely transmit the quantum state from the classical node to the QPU, applying a prepare-and-send methodology to preserve data blindness during execution. Finally, the resulting quantum measurements are returned to the classical cloud for post-processing and aggregation before the final analysis is delivered to the end user. The core design choice in this approach is the adaptation of selective scheduling heuristics to manage the delicate interaction between classical datacenters and quantum nodes. Traditional task allocation algorithms are augmented to treat QPUs as highly specialized, time-shared resources with exceptionally strict execution windows dictated by qubit decoherence times. Furthermore, we utilize an ad hoc resource harvesting concept for the classical preprocessing stages, allowing sporadic, non-exclusive classical resources to rapidly feed structured data into the highly controlled quantum environment. This strict division of labor ensures that the expensive, high-maintenance quantum hardware is never left idle waiting for classical data transformations, thereby optimizing the overall makespan and mitigating critical performance bottlenecks. To validate the efficacy of this proposed framework, we outline a comprehensive hypothetical evaluation plan. We propose extending a standard cloud simulation environment, such as CloudSim, by integrating a custom quantum execution module that mathematically models qubit gate latencies, state preparation times, and network communication overheads. The hypothetical evaluation will benchmark our hybrid scheduling algorithm against strict First-Come-First-Serve (FCFS) provisioning policies across varying data analysis workload intensities. By tracking the total task makespan and server utilization rates, we anticipate demonstrating that intelligent algorithmic partitioning significantly reduces the performance degradation typically associated with delegated quantum computing overheads. 4.Practical implications The deployment of hybrid quantum-cloud systems carries significant practical implications for the future of enterprise IT infrastructure. From a deployment perspective, organizations will no longer need to invest heavily in fragile, independent quantum hardware; instead, they will consume computation through standardized cloud providers. This shift mandates a reevaluation of existing academic cloud research, pushing industry professionals to move beyond small-scale toy deployments and embrace complex, data-center-level abstractions that in

Jun 14, 20250 citations

Classical-To-Quantum Transfer Learning for Spoken Command Recognition Based on Quantum Neural Networks

Jun Qi, Javier Tejedor

This work investigates an extension of transfer learning applied in machine learning algorithms to the emerging hybrid end-to-end quantum neural network (QNN) for spoken command recognition (SCR). Our QNN-based SCR system is composed of classical and quantum components: (1) the classical part mainly relies on a 1D convolutional neural network (CNN) to extract speech features; (2) the quantum part is built upon the variational quantum circuit with a few learnable parameters. Since it is inefficient to train the hybrid end-to-end QNN from scratch on a noisy intermediate-scale quantum (NISQ) device, we put forth a hybrid transfer learning algorithm that allows a pre-trained classical network to be transferred to the classical part of the hybrid QNN model. The pre-trained classical network is further modified and augmented through jointly fine-tuning with a variational quantum circuit (VQC). The hybrid transfer learning methodology is particularly attractive for the task of QNN-based SCR because low-dimensional classical features are expected to be encoded into quantum states. We assess the hybrid transfer learning algorithm applied to the hybrid classical-quantum QNN for SCR on the Google speech command dataset, and our classical simulation results suggest that the hybrid transfer learning can boost our baseline performance on the SCR task.

IEEE International Conference on Acoustics Speech and Signal ProcessingApr 27, 202250 citations

The Role of Quantum in Hybrid Quantum-Classical Neural Networks: An Empirical Assessment

Dominik Freinberger, Philipp Moser

Quantum machine learning has emerged as a promising application domain for near-term quantum hardware, particularly through hybrid quantum-classical models that leverage both classical and quantum processing. Although numerous hybrid architectures have been proposed and demonstrated successfully on benchmark tasks, a significant open question remains regarding the specific contribution of quantum components to the overall performance of these models. In this work, we aim to shed light on the impact of quantum processing within hybrid quantum-classical neural network architectures through a rigorous statistical study under ideal, noise-free simulation. We systematically assess a selection of commonly used HQNN architectures on three binary medical classification tasks covering medical signal data as well as planar and volumetric images. In doing so, we examine the influence attributable to classical and quantum aspects such as encoding schemes, entanglement, and circuit size. Within this experimental setting, we find that in best-case scenarios, hybrid models show performance comparable to their classical counterparts. However, in most cases, performance metrics deteriorate when the investigated QNN components replace matched classical processing layers. Our multi-modal analysis provides empirical insights into the contributions of the tested quantum components and advocates for cautious claims and design choices for hybrid models in near-term applications.

quant-pharXiv:2601.04732Jan 9, 20261 citation