Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE Transactions on Mobile Computing

A Quantum-Driven Efficient Learning Model for Enhancing Robustness of IoT Topology

Songwei Zhang, Tie Qiu, Xiaobo Zhou, Yusheng Ji

Published Jul 8, 2025• 1 citation
Publisher

Abstract

The robustness of Internet of Things (IoT) topologies measures a network structure's tolerance to random failures, or attacks, which is crucial for stable network communication. Research on optimizing network topology robustness has shifted from empirical rules and heuristics to machine learning, which can extract the features of robust network from topology data, thereby reducing the complexity of traditional topology optimization. However, machine learning approaches typically require a large number of parameters, resulting in high costs associated with parameter tuning and inference. To address these issues, this paper combines parameterized quantum circuits, and proposes a Quantum-Driven efficient Learning Model (QDLM) for enhancing robustness of IoT topology. This model leverages quantum exponential states to significantly reduce the number of training parameters while preserving learning performance. For inputs, QDLM integrates arithmetic encoding and quantum state encoding based on topological adjacency matrix, reducing the number of neurons. In training phase, parameterized quantum rotation gates and controlled quantum gates are used to achieve efficient training. A quantum measurement method is designed to ensure the output topology is a connected graph with the required number of edges. Compared to existing topology learning models, QDLM achieves an order-of-magnitude reduction in training parameters while maintaining topology learning effectiveness.

DOI
10.1109/tmc.2025.3586749

Related papers

Machine Learning Assisted Quantum Network Optimization for Future Internet Architectures

Ankur Kumar, Harish Dutt Sharma, Pintu Kumar

The integration of quantum networking with future Internet architectures introduces new challenges in resource optimization, routing, and network reliability. Traditional optimization techniques are often insufficient to handle the dynamic and probabilistic nature of quantum networks. This paper proposes a machine learning-assisted framework for quantum network optimization, where data-driven models are used to enhance decision-making in routing, resource allocation, and entanglement management. The approach combines classical learning algorithms with quantum network control mechanisms to improve adaptability and performance under varying network conditions. Experimental evaluation demonstrates improved efficiency in network utilization, reduced latency, and enhanced robustness compared to conventional methods. The results highlight the potential of machine learning-assisted optimization as a key enabler for scalable and intelligent quantum Internet infrastructures.

International Journal of Engineering Technology and Computer Research•Apr 30, 2026•0 citations

Quantum Topology-Driven Machine Learning

Jincheng Zhang

Quantum Topology-Driven Machine Learning explores the potential of combining quantum topology theory with machine learning to enhance model generalization. This paper investigates how utilizing quantum topology as a learning objective can improve model robustness and performance. We outline the core mechanism, detailing how quantum entanglement and topological phases are leveraged to create novel learning paradigms. The research aims to demonstrate a shift from traditional data-dependent models towards a more adaptable and potentially more powerful approach, specifically focusing on enhancing the generalization capabilities of machine learning algorithms. The paper concludes by discussing future research directions and potential applications within the field of artificial intelligence.

•Sep 3, 2026•0 citations

IPQC: An Intelligent Quantum Graph Convolutional Network for Topological Data Processing on Green IoT

Naixue N. Xiong, Silong Li, Linshu Chen +2 more

Processing graph-structured data in Green Internet of Things (IoT) applications demands a dual focus on analytical accuracy and energy efficiency. While Quantum Graph Neural Networks (QGNNs) present a promising computational paradigm, the parameterized quantum circuits (PQC) they rely on often require excessive depth and lack robustness, hindering their use in resource-constrained environments. The core problem addressed in this paper is how to design a quantum-inspired GCN that balances expressive power with shallow, energy-efficient circuit design, making it viable for sustainable Green IoT deployment. Our main contributions are threefold. First, we design the Intelligent Parameterized Quantum Circuit (IPQC) as a compact, 15-parameter bidirectional-control quantum convolutional block that enhances expressive power while maintaining parameter efficiency and noise tolerance. Second, building on this block, we develop the Quantum Graph Convolutional Network with Residual Injection (QGCN-RI), a residual-injection-driven network that integrates two-stage normalization and amplitude encoding to significantly improve optimization stability. Third, we conduct comprehensive experiments on citation network benchmarks. Results demonstrate that QGCN-RI achieves performance competitive with strong classical baselines like GAT, reaching 83.3% accuracy on Cora. More critically, our quantitative analysis indicates that the model’s shallow circuit design allows it to achieve this competitive accuracy with a lower estimated energy consumption. By showing that a compact QGCN can approximate the performance of its classical counterparts with reduced resource costs, our work validates the feasibility of quantum graph learning for developing sustainable and resource-aware Green IoT solutions.

IEEE Transactions on Green Communications and Networking•Oct 28, 2025•0 citations

TinyQL: A Quantum Machine Learning Framework at Edge for Resource-Constrained IoT Devices

Maurice Ngouen, Mohammad Ashiqur Rahman, Alexander Perez-Pons +1 more

Implementing efficient machine learning (ML) is extremely difficult for Internet of Things (IoT) devices with limited computational power, memory, and energy resources. We introduce TinyQL, a quantum machine learning framework for Internet of Things (IoT) devices with limited resources. The framework tackles three main issues: dimensionality reduction using variational quantum circuits while maintaining anomaly detection capabilities; model compression using entropy-based structured pruning and parameter quantization; and effective encoding of classical network traffic data into quantum states using amplitude embedding. TinyQL can be deployed on devices with limited quantum simulation capabilities by combining a quantum autoencoder with traditional optimization approaches. The CICIDS 2017 intrusion detection dataset is used to assess the framework, and it is contrasted with traditional autoencoder methods. According to experimental data, TinyQL requires substantially fewer parameters and has a smaller memory footprint than classical approaches without degradation of detection accuracy. Deployment on resource-constrained IoT devices is made possible by the optimized models’ $\mathbf{9 8. 0 8 \%}$ detection accuracy and up to 93.75% memory savings when compared to the original quantum model. An entropy-based pruning strategy tailored for quantum circuits, a novel quantum autoencoder architecture optimized for IoT constraints, a thorough evaluation framework contrasting quantum and classical approaches for IoT anomaly detection, and useful recommendations for implementing quantum machine learning models on edge devices are among the contributions.

International Symposium on Quality Electronic Design (ISQED)•Apr 8, 2026•0 citations

Quantum Machine Learning Topology Modeling

Jincheng Zhang

Quantum Machine Learning Topology Modeling explores the potential of leveraging the topological structure of quantum systems to model and optimize machine learning models. This paper proposes a novel framework that utilizes topological properties to represent and enhance the learning process of machine learning algorithms. We introduce a quantum machine learning model that dynamically adapts to the inherent topology of the data, leading to improved generalization and parameter optimization. The core mechanism centers on identifying and exploiting key topological features within datasets, translating these features into quantum operations that guide the learning process. We demonstrate the effectiveness of this approach through simulations and preliminary experiments, showcasing the potential for improved model performance and robustness.

•Sep 2, 2026•0 citations

Quantum-Enhanced IoT-Cloud Security: Integrating SHAP and Variational Quantum Classifiers

Veena Antony, Veena Antony, Nainan Thangarasu

In the current Internet era, there are now trillions of gadgets online, and the Internet of Things is becoming a necessary part of daily life. IoT devices are connected, but this also makes them vulnerable to cyberattacks. Cyberattacks targeting Internet of Things (IoT) systems have increased dramatically in both volume and sophistication in the last year. Determining the importance and explainability of significant feature selection is not done in conventional feature selection, which acts as a black box method. Classical machine learning suffers from overload and class imbalance issues in IoT-based cloud security which is the major issue that results in botnet attack detection. To overcome these two issues, this study developed a model of a feature map to encode conventional data to a quantum feature space and then utilize the newly created quantum data in the cognitive circuit, which motivates to development of Quantum Shapley Additive Explanation with Variational Quantum Classifier (QSHAP-VQC) is implemented. This makes it possible to employ classical data in a quantum circuit. To minimize a cost function, a VQC employs hybrid quantum-classical techniques that involve parameterized circuits and gates whose parameters are improved via a classically based optimization loop. A quantum-classical hybrid loop consisting of these steps is eventually broken when the classical optimization finds the ideal parameters. For training data, the usual cost function is a comparison of the actual and expected outputs. The proposed QSHAP-VQC achieves the highest rate of accuracy in the detection of attacks in an IoT cloud environment.

Journal of Computer Science•Jan 1, 2025•1 citation
The robustness of Internet of Things (IoT) topologies measures a network structure's tolerance to random failures, or attacks, which is crucial for stable network communication. Research on optimizing network topology robustness has shifted from empirical rules and heuristics to machine learning, which can extract the features of robust network from topology data, thereby reducing the complexity of traditional topology optimization. However, machine learning approaches typically require a large number of parameters, resulting in high costs associated with parameter tuning and inference. To address these issues, this paper combines parameterized quantum circuits, and proposes a Quantum-Driven efficient Learning Model (QDLM) for enhancing robustness of IoT topology. This model leverages quantum exponential states to significantly reduce the number of training parameters while preserving learning performance. For inputs, QDLM integrates arithmetic encoding and quantum state encoding based on topological adjacency matrix, reducing the number of neurons. In training phase, parameterized quantum rotation gates and controlled quantum gates are used to achieve efficient training. A quantum measurement method is designed to ensure the output topology is a connected graph with the required number of edges. Compared to existing topology learning models, QDLM achieves an order-of-magnitude reduction in training parameters while maintaining topology learning effectiveness.
The integration of quantum networking with future Internet architectures introduces new challenges in resource optimization, routing, and network reliability. Traditional optimization techniques are often insufficient to handle the dynamic and probabilistic nature of quantum networks. This paper proposes a machine learning-assisted framework for quantum network optimization, where data-driven models are used to enhance decision-making in routing, resource allocation, and entanglement management. The approach combines classical learning algorithms with quantum network control mechanisms to improve adaptability and performance under varying network conditions. Experimental evaluation demonstrates improved efficiency in network utilization, reduced latency, and enhanced robustness compared to conventional methods. The results highlight the potential of machine learning-assisted optimization as a key enabler for scalable and intelligent quantum Internet infrastructures.
Quantum Topology-Driven Machine Learning explores the potential of combining quantum topology theory with machine learning to enhance model generalization. This paper investigates how utilizing quantum topology as a learning objective can improve model robustness and performance. We outline the core mechanism, detailing how quantum entanglement and topological phases are leveraged to create novel learning paradigms. The research aims to demonstrate a shift from traditional data-dependent models towards a more adaptable and potentially more powerful approach, specifically focusing on enhancing the generalization capabilities of machine learning algorithms. The paper concludes by discussing future research directions and potential applications within the field of artificial intelligence.
Processing graph-structured data in Green Internet of Things (IoT) applications demands a dual focus on analytical accuracy and energy efficiency. While Quantum Graph Neural Networks (QGNNs) present a promising computational paradigm, the parameterized quantum circuits (PQC) they rely on often require excessive depth and lack robustness, hindering their use in resource-constrained environments. The core problem addressed in this paper is how to design a quantum-inspired GCN that balances expressive power with shallow, energy-efficient circuit design, making it viable for sustainable Green IoT deployment. Our main contributions are threefold. First, we design the Intelligent Parameterized Quantum Circuit (IPQC) as a compact, 15-parameter bidirectional-control quantum convolutional block that enhances expressive power while maintaining parameter efficiency and noise tolerance. Second, building on this block, we develop the Quantum Graph Convolutional Network with Residual Injection (QGCN-RI), a residual-injection-driven network that integrates two-stage normalization and amplitude encoding to significantly improve optimization stability. Third, we conduct comprehensive experiments on citation network benchmarks. Results demonstrate that QGCN-RI achieves performance competitive with strong classical baselines like GAT, reaching 83.3% accuracy on Cora. More critically, our quantitative analysis indicates that the model’s shallow circuit design allows it to achieve this competitive accuracy with a lower estimated energy consumption. By showing that a compact QGCN can approximate the performance of its classical counterparts with reduced resource costs, our work validates the feasibility of quantum graph learning for developing sustainable and resource-aware Green IoT solutions.
Implementing efficient machine learning (ML) is extremely difficult for Internet of Things (IoT) devices with limited computational power, memory, and energy resources. We introduce TinyQL, a quantum machine learning framework for Internet of Things (IoT) devices with limited resources. The framework tackles three main issues: dimensionality reduction using variational quantum circuits while maintaining anomaly detection capabilities; model compression using entropy-based structured pruning and parameter quantization; and effective encoding of classical network traffic data into quantum states using amplitude embedding. TinyQL can be deployed on devices with limited quantum simulation capabilities by combining a quantum autoencoder with traditional optimization approaches. The CICIDS 2017 intrusion detection dataset is used to assess the framework, and it is contrasted with traditional autoencoder methods. According to experimental data, TinyQL requires substantially fewer parameters and has a smaller memory footprint than classical approaches without degradation of detection accuracy. Deployment on resource-constrained IoT devices is made possible by the optimized models’ 98.08%\mathbf{9 8. 0 8 \%}98.08% detection accuracy and up to 93.75% memory savings when compared to the original quantum model. An entropy-based pruning strategy tailored for quantum circuits, a novel quantum autoencoder architecture optimized for IoT constraints, a thorough evaluation framework contrasting quantum and classical approaches for IoT anomaly detection, and useful recommendations for implementing quantum machine learning models on edge devices are among the contributions.
Quantum Machine Learning Topology Modeling explores the potential of leveraging the topological structure of quantum systems to model and optimize machine learning models. This paper proposes a novel framework that utilizes topological properties to represent and enhance the learning process of machine learning algorithms. We introduce a quantum machine learning model that dynamically adapts to the inherent topology of the data, leading to improved generalization and parameter optimization. The core mechanism centers on identifying and exploiting key topological features within datasets, translating these features into quantum operations that guide the learning process. We demonstrate the effectiveness of this approach through simulations and preliminary experiments, showcasing the potential for improved model performance and robustness.
In the current Internet era, there are now trillions of gadgets online, and the Internet of Things is becoming a necessary part of daily life. IoT devices are connected, but this also makes them vulnerable to cyberattacks. Cyberattacks targeting Internet of Things (IoT) systems have increased dramatically in both volume and sophistication in the last year. Determining the importance and explainability of significant feature selection is not done in conventional feature selection, which acts as a black box method. Classical machine learning suffers from overload and class imbalance issues in IoT-based cloud security which is the major issue that results in botnet attack detection. To overcome these two issues, this study developed a model of a feature map to encode conventional data to a quantum feature space and then utilize the newly created quantum data in the cognitive circuit, which motivates to development of Quantum Shapley Additive Explanation with Variational Quantum Classifier (QSHAP-VQC) is implemented. This makes it possible to employ classical data in a quantum circuit. To minimize a cost function, a VQC employs hybrid quantum-classical techniques that involve parameterized circuits and gates whose parameters are improved via a classically based optimization loop. A quantum-classical hybrid loop consisting of these steps is eventually broken when the classical optimization finds the ideal parameters. For training data, the usual cost function is a comparison of the actual and expected outputs. The proposed QSHAP-VQC achieves the highest rate of accuracy in the detection of attacks in an IoT cloud environment.