It seems as though progressively more people are in the race to upload content, data, and information online; and hospitals haven't neglected this trend either. Hospitals are now at the forefront for ...multi-site medical data sharing to provide ground-breaking advancements in the way health records are shared and patients are diagnosed. Sharing of medical data is essential in modern medical research. Yet, as with all data sharing technology, the challenge is to balance improved treatment with protecting patient's personal information. This paper provides a novel split learning algorithm coined the term, "multi-site split learning", which enables a secure transfer of medical data between multiple hospitals without fear of exposing personal data contained in patient records. It also explores the effects of varying the number of end-systems and the ratio of data-imbalance on the deep learning performance. A guideline for the most optimal configuration of split learning that ensures privacy of patient data whilst achieving performance is empirically given. We argue the benefits of our multi-site split learning algorithm, especially regarding the privacy preserving factor, using CT scans of COVID-19 patients, X-ray bone scans, and cholesterol level medical data.
In many deep neural network (DNN) applications, the difficulty of gathering high‐quality data in industry fields hinders the practical use of DNN. Thus, the concept of transfer learning (TL) has ...emerged, which leverages the pretrained knowledge of the DNN which was built based on large‐scale datasets. For this TL objective, this paper suggests two‐stage architectural fine‐tuning for reducing the costs and time while exploring the most efficient DNN model, inspired by neural architecture search (NAS). The first stage is mutation, which reduces the search costs using a priori architectural information. Moreover, the next stage is early‐stopping, which reduces NAS costs by terminating the search process in the middle of computation. The data‐intensive experimental results verify that the proposed method outperforms benchmarks.
This paper suggests two‐stage architectural fine‐tuning for reducing the costs and time while exploring the most efficient neural network model, inspired by neural architecture search (NAS). The first stage is mutation, which reduces the search costs using a priori architectural information. Moreover, the next stage is early‐stopping, which reduces NAS costs by terminating the search process in the middle of computation.
Green tide, which is a serious water pollution problem, is caused by the complex relationships of various factors, such as flow rate, several water quality indicators, and weather. Because the ...existing methods are not suitable for identifying these relationships and making accurate predictions, a new system and algorithm is required to predict the green tide phenomenon and also minimize the related damage before the green tide occurs. For this purpose, we consider a new network model using smart sensor-based federated learning which is able to use distributed observation data with geologically separated local models. Moreover, we design an optimal scheduler which is beneficial to use real-time big data arrivals to make the overall network system efficient. The proposed scheduling algorithm is effective in terms of (1) data usage and (2) the performance of green tide occurrence prediction models. The advantages of the proposed algorithm is verified via data-intensive experiments with real water quality big-data.
Recent advances in low Earth orbit (LEO) satellites have made it possible to achieve zero blind spots on Earth. Considering the give locations of these devices, this makes satellite-ground links ...between quantum devices a practical possibility. This paper proposes the first quantum federated learning (QFL) application in satellite-ground communication. To improve communication and computing performance, this paper adopts slimmable quantum federated learning (SQFL) and slimmable quantum neural networks (sQNN), which allow for two different configurations in quantum neural networks: the angle and pole configurations. This paper also employs superposition coding and successive decoding to increase communication opportunities. Through extensive experiments, the proposed satellite-ground SQFL framework performs well and is both computationally and communicationally efficient compared to classical federated learning and QFL.
Cloud computing has revolutionized information technology (IT) infrastructure by providing benefits such as scalability, flexibility, and convenience. If the radio access networks of wireless systems ...are constructed on edge clouds, operators can obtain various advantages. One notable benefit is pooling gain achieved through auto-scaling, leading to savings on investment. However, conventional auto-scaling methods employ numerous instances to prevent traffic overloads and ensure reliable system performance during peak demand. This traditional approach increases the burden on computing resources. To address this, this paper introduces an intelligent auto-scaling scheme that enhances pooling gain by leveraging the distinctive attributes of cloud computing. Additionally, we refine pooling gain to measure the performance of the fifth generation (5G) base stations on edge clouds. The simulation outcomes of our advanced auto-scaling scheme demonstrate its ability to achieve substantial improvements in productivity compared to the simpler implementations used in distributed cell sites, both in terms of pooling gain and the efficient utilization of computing resources.
The urban aerial mobility (UAM) system, such as drone taxi or air taxi, is one of future on-demand transportation networks. Among them, electric vertical takeoff and landing (eVTOL) is one of UAM ...systems that is for identifying the locations of passengers, flying to the positions where the passengers are located, loading the passengers, and delivering the passengers to their destinations. In this paper, we propose a distributed deep reinforcement learning where the agents are formulated as eVTOL vehicles that can compute the optimal passenger transportation routes under the consideration of passenger behaviors, collisions among eVTOL, and eVTOL battery status.
In modern networking research, infrastructure-assisted unmanned autonomous vehicles (UAVs) are actively considered for real-time learning-based surveillance and aerial data-delivery under unexpected ...3D free mobility and coordination. In this system model, it is essential to consider the power limitation in UAVs and autonomous object recognition (for abnormal behavior detection) deep learning model performance in infrastructure/towers. To overcome the power limitation of UAVs, this paper proposes a novel aerial scheduling algorithm between multi-UAVs and multi-towers where the towers conduct wireless power transfer toward UAVs. In addition, to take care of the high-performance learning model training in towers, the training data from UAVs should be fairly delivered to towers because larger data delivery will make huge computation overhead whereas the less data delivery will introduce overfitting. Therefore, fair data distribution from UAVs to their associated towers is essentially required. Therefore, this paper proposes a novel workload-aware scheduling algorithm between multi-towers and multi-UAVs for joint power-charging from towers to their associated UAVs and training data delivery from UAVs to their associated towers. For computing the workload-aware optimal scheduling decisions in each unit time, our solution approach for the given scheduling problem is designed based on Markov decision process (MDP) to deal with (i) time-varying low-complexity computation and (ii) pseudo-polynomial optimality. As shown in performance evaluation results, our proposed algorithm ensures (i) sufficient times for resource exchanges between towers and UAVs, (ii) the most even and uniform data collection during the processes compared to the other algorithms, and (iii) the performance of all towers converges to optimal levels.
As one of the latest fields of interest in both academia and industry, quantum computing has garnered significant attention. Among various topics in quantum computing, variational quantum circuits ...(VQCs) have been noticed for their ability to carry out quantum deep reinforcement learning (QRL). This article verifies the potential of QRL, which will be further realized by implementing quantum multiagent reinforcement learning (QMARL) from QRL, especially for Internet-connected autonomous multirobot control and coordination in smart factory applications. However, the extension is not straightforward due to the nonstationarity of classical MARL. To cope with this, the centralized training and decentralized execution (CTDE) QMARL framework is proposed under the Internet connection. A smart factory environment with the Internet of Things (IoT)-based multiple agents is used to show the efficacy of the proposed algorithm. The simulation corroborates that the proposed QMARL-based autonomous multirobot control and coordination performs better than the other frameworks.
This paper surveys recent multiagent reinforcement learning and neural Myerson auction deep learning efforts to improve mobility control and resource management in autonomous ground and aerial ...vehicles. The multiagent reinforcement learning communication network (CommNet) was introduced to enable multiple agents to perform actions in a distributed manner to achieve shared goals by training all agents' states and actions in a single neural network. Additionally, the Myerson auction method guarantees trustworthiness among multiple agents to optimize rewards in highly dynamic systems. Our findings suggest that the integration of MARL CommNet and Myerson techniques is very much needed for improved efficiency and trustworthiness.
The electric vehicle (EV) market increases due to the benefits of reducing greenhouse gas emissions using renewable energy resources. In this context, the charging scheme of electric vehicles in ...charging stations (CSs) is also important. Electronic devices’ charging between EV and multiple CS should consider EV’s short battery capacity, long charging time, residual energy in each CS, and time of use (ToU) for charging. In this paper, multiple CSs compete to offer electricity charging to a single EV. Based on this need, this paper proposes a deep learning-based auction which increases the charging amounts using Myerson auction while preserving truthfulness.
PDF
