In this era of automation, drones have considered to be a key element with its wide applicability in various situations. Additionally, the efficiency of drones can be further improved via collaborated learning techniques such as federated learning (FL). However, the drone-based FL requires a significant amount of communications because the drones share their FL data over wireless channels. Therefore, the unstable channel condition of wireless transmission introduces the low success probability in large-scale parameter transmission. According to quantum computing (QC) advantages, this problem can be tackled due to less parameter utilization in quantum neural network (QNN) training. Therefore, a novel QC-based FL known as theadaptive quantum federated learning (AQFL)is proposed for drone license plate recognition in autonomous surveillance applications. In our proposed AQFL, adaptive QNNs (AQNNs) are used as local models which can control the depths of QNNs by adding measurement computations in the middles of QNN architectures. Thus, our proposed AQFL is robust even in training data non-iidness and large local model variances while maintaining sufficient learning performance. Finally, the evaluation results verify that our proposed AQFL achieves the desired performance improvements.
In this era of automation, drones have considered to be a key element with its wide applicability in various situations. Additionally, the efficiency of drones can be further improved via collaborated learning techniques such as federated learning (FL). However, the drone-based FL requires a significant amount of communications because the drones share their FL data over wireless channels. Therefore, the unstable channel condition of wireless transmission introduces the low success probability in large-scale parameter transmission. According to quantum computing (QC) advantages, this problem can be tackled due to less parameter utilization in quantum neural network (QNN) training. Therefore, a novel QC-based FL known as theadaptive quantum federated learning (AQFL)is proposed for drone license plate recognition in autonomous surveillance applications. In our proposed AQFL, adaptive QNNs (AQNNs) are used as local models which can control the depths of QNNs by adding measurement computations in the middles of QNN architectures. Thus, our proposed AQFL is robust even in training data non-iidness and large local model variances while maintaining sufficient learning performance. Finally, the evaluation results verify that our proposed AQFL achieves the desired performance improvements.
In recent years, unmanned aerial vehicles (UAVs) have proven their effectiveness in surveillance due to their superior mobility. By utilizing multiple UAVs with collaborated learning, surveillance of a huge area while consuming minimum resources is possible. However, the performance of UAV collaborative learning systems is still severely limited because the number of parameters in a classical NN is generally very large, which is unsuitable for unstable wireless communication. To address this issue, this paper uses a quantum neural network (QNN), which has large computational capabilities and uses fewer parameters to overcome the problems caused by many parameters. However, it is still difficult to send all the parameters of the QNN to the server under poor channel conditions. Therefore, this paper proposes dynamic quantum federated learning (DQFL), a novel framework designed for UAVs employing quantum computing (QC) and federated learning (FL). The proposed DQFL uses a dynamic quantum neural network (DQNN) with a multi-depth circuit and employs dynamic control of the circuit layer to improve the efficiency of local parameter transmission between UAVs in unstable wireless communication environments. Extensive simulations conducted under real-world autonomous surveillance conditions demonstrate the robustness of DQFL to non-iidness, varying signal-to-noise ratios (SNRs), and poor communication channel conditions in the UAV environment.
With the rise of autonomous vehicles (AVs), the number of self-driving vehicles has increased in recent years and vehicular computing (VC) was actively used to ensure road safety by controlling AVs. However, the recent escalation in demand for a massive scale vehicle network has prevented VC from achieving its purpose. To overcome this problem, a framework which combined VC and federated learning (FL) was proposed. Although the vehicular FL was initially effective, it also faced challenges as the excessive increase in the amount and size of data generated by AVs and the serious threat of data leakage have severely degraded the performance of vehicular FL. Thus, quantum computing was exploited to further improve on the pre-existing FL and proposed dynamic quantum federated learning (DQFL). This work proposes the application of DQFL to public safety scenarios to investigate the effectiveness of the model in controlling the AVs under real-life constraints.
Recent advancements in Quantum Neural Networks (QNNs) have demonstrated theoretical and experimental performance superior to their classical counterparts in a wide range of applications. However, existing centralized QNNs cannot solve many real-world problems because collecting large amounts of training data to a common public site is time-consuming and, more importantly, violates data privacy. Federated Learning (FL) is an emerging distributed machine learning framework that allows collaborative model training on decentralized data residing on multiple devices without breaching data privacy. Some initial attempts at Quantum Federated Learning (QFL) either only focus on improving the QFL performance or rely on a trusted quantum server that fails to preserve data privacy. In this work, we propose CryptoQFL, a QFL framework that allows distributed QNN training on encrypted data. CryptoQFL is (1) secure, because it allows each edge to train a QNN with local private data, and encrypt its updates using quantum homomorphic encryption before sending them to the central quantum server; (2) communication-efficient, as CryptoQFL quantize local gradient updates to ternary values, and only communicate non-zero values to the server for aggregation; and (3) computation-efficient, as CryptoQFL presents an efficient quantum aggregation circuit with significantly reduced latency compared to state-of-the-art approaches.
In recent years, the growing demand for secure, efficient, and scalable machine learning has highlighted significant challenges in traditional Federated Learning (FL) systems. These challenges include computational inefficiencies in handling large-scale, high-dimensional data, privacy concerns, and the limitations of classical hardware in solving complex optimization problems. Additionally, the increasing complexity of data and the need for more robust privacy-preserving methods have pushed the boundaries of classical FL. To address these issues, this paper presents a detailed framework for Quantum Federated Learning (QFL), where quantum computing and FL converge to enable distributed quantum model training with enhanced privacy preservation. The system is modeled using local quantum models on clients and a global aggregation strategy at the server. We develop quantum gradient-based optimization with quantum cross-entropy loss, privacy preservation through encryption, and robustness to quantum noise. A set of theorems is provided to prove the correctness and efficiency of the model, supported by rigorous mathematical formulations. We used IBM Qiskit for the simulation and compared our proposed QFL with classical FL. The results demonstrate that our QFL outperformed classical FL in terms of accuracy, precision, recall, and F1 score.
Federated learning (FL) is a privacy-preserving alternative to centralized machine learning, where model training is performed on local devices and only global model updates are shared, effectively addressing challenges, such as data silos and privacy protection. Recently, quantum FL (QFL), an emerging FL branch, has garnered significant attention in many industry applications. However, existing QFL algorithms predominantly employ average weighting for global model training, which shows poor performance on heterogeneous healthcare data. To address this challenge, this study proposes a dynamic aggregation QFL algorithm (DAQFL) for intelligent diagnosis. Specifically, it utilizes quantum neural networks (QNNs) as local training models and designs corresponding variational quantum circuits (VQC). To mitigate performance degradation caused by the heterogeneity of medical industrial data, a dynamic aggregation method based on accuracy is proposed to enhance global model performance effectively. Extensive experiments with three distribution settings, including independent and identically distributed (IID), non-independent and identically distributed (Non-IID), and long-tail datasets, show that DAQFL outperforms baseline algorithms in accuracy and training speed. It also performs well in privacy protection and robustness of anti-noise, improving its suitability for real-world medical applications.
Upon integrating Quantum Neural Network (QNN) as the local model, Quantum Federated Learning (QFL) has recently confronted notable challenges. Firstly, exploration is hindered over sharp minima, decreasing learning performance. Secondly, the steady gradient descent results in more stable and predictable model transmissions over wireless channels, making the model more susceptible to attacks from adversarial entities. Additionally, the local QFL model is vulnerable to noise produced by the quantum device's intermediate noise states, since it requires the use of quantum gates and circuits for training. This local noise becomes intertwined with learning parameters during training, impairing model precision and convergence rate. To address these issues, we propose a new QFL technique that incorporates differential privacy and introduces a dedicated noise estimation strategy to quantify and mitigate the impact of intermediate quantum noise. Furthermore, we design an adaptive noise generation scheme to alleviate privacy threats associated with the vanishing gradient variance phenomenon of QNN and enhance robustness against device noise. Experimental results demonstrate that our algorithm effectively balances convergence, reduces communication costs, and mitigates the adverse effects of intermediate quantum noise while maintaining strong privacy protection. Using real-world datasets, we achieved test accuracy of up to 98.47% for the MNIST dataset and 83.85% for the CIFAR-10 dataset while maintaining fast execution times.