Traditional centralized model training assumes that data samples are readily available and can be processed without constraints. In contrast, decentralized machine learning (DML) addresses the limitation by collaborative model training and inference directly on distributed data sources. The transformation from data centralization to decentralization helps comply with data regulations and improves system scalability with reduced reliance on cloud servers. However, a tradeoff between model personalization and generalization exists: the fine-tuning of local training data distribution sacrifices model generalization on the testing data distribution that differs from the training data distribution. To improve the tradeoff, we propose a DML framework that can inherently make model personalization and generalization easier by selecting a model among multiple ones judiciously. We develop a scalable selector for model selection and use blockchain to achieve model consensus. The personalized model selector is then proposed for test-time adaptation. Using computer simulations, we show that our method not only outperforms competitive personalization benchmarks but also generalizes well for new data distributions with various shifts.

In the Internet of Things (IoT) employing centralized machine learning, security is a major concern due to the heterogeneity of end devices. Malicious devices could launch poisoning attacks to degrade machine learning models. Distributed machine learning (DML) with blockchain provides a potential solution. Once local weights are recorded on the blockchain, model aggregation with defensive schemes can be executed on smartphones to prevent attacks. However, blockchain with the proof-of-work (PoW) consensus mechanism wastes computing resources and adds latency to DML. Computing resources can be utilized more efficiently with proof-of-useful-work (uPoW), which secures transactions by solving relevant real-world problems. We propose a novel uPoW method to minimize per-round latency of DML. The uPoW mining process schedules DML instances among multi-access edge computing (MEC) servers by solving a multi-way number partitioning problem. Moreover, poisoning attacks on heterogeneous training data pose significant challenges to blockchain-based DML. To address this problem, we propose a novel aggregation protocol, named$\mathit{Corrected Krum}$, to counter such attacks and improve the convergence speed of DML. By leveraging the mean-field approximation method, training errors are corrected to reduce the negative impact of poisoning attacks. Simulation results show that our proposed blockchain approach can significantly speed up DML compared with benchmarks.

In Internet of Things (IoT) employing centralized machine learning, security is a major concern due to the heterogeneity of end devices. Decentralized machine learning (DML) with blockchain is a potential solution. However, blockchain with proof-of-work (PoW) consensus mechanism wastes computing resources and adds latency to DML. Computing resources can be utilized more efficiently with proof-of-useful-work (uPoW), which secures transactions by solving real-world problems. We propose a novel uPoW method that exploits PoW mining to accelerate DML through a task scheduling framework for multi-access edge computing (MEC) systems. To provide a good quality-of-service for the system, we minimize the latency by solving a multi-way number partitioning problem in the extended form. A novel uPoW-based mechanism is proposed to schedule DML tasks among MEC servers effectively. Simulation results show that our proposed blockchain strategies accelerate DML significantly compared with benchmarks.

The use of flying platforms such as unmanned aerial vehicles (UAVs), popularly known as drones, is rapidly growing. UAVs can greatly support data collecting and processing for Internet of Things devices (IoTDs) in mobile edge computing (MEC) systems due to their advantages of high environmental flexibility. This paper focuses on the scenario where multiple heterogeneous rotary-wing UAVs complete data collection and processing missions cooperatively. This paper introduces an energy minimization problem for UAV-assisted MEC system which attempts to optimize route planning and task allocation of UAVs. The energy consumption of a UAV includes hovering energy and flight energy depending on its configuration. By jointly choosing optimal UAVs for tasks and routes, we aim to obtain a sub-optimal solution of allocating IoTD tasks to UAVs and UAV flying route design while minimizing energy consumption. The Ant Colony System (ACS) algorithm is employed to obtain a high-quality near-optimal solution to solve this optimization problem. Finally, the simulation results show the effectiveness and efficiency of our proposed solution.

This paper considers a multi-drone enabled data collection system for smart cities, where there are two kinds of drones, i.e., Low Altitude Platforms (LAPs) and a High Altitude Platform (HAP). In the proposed system, the LAPs perform data collection tasks for smart cities and the solar-powered HAP provides energy to the LAPs using wireless laser beams. We aim to minimize the total laser charging energy of the HAP, by jointly optimizing the LAPs' trajectory and the laser charging duration for each LAP, subject to the energy capacity constraints of the LAPs. This problem is formulated as a mixed-integer and non-convex Drones Traveling Problem (DTP), which is a combinatorial optimization problem and NP-hard. We propose an efficient and novel search algorithm named Drones Traveling Algorithm (DTA) to obtain a near-optimal solution. Simulation results show that DTA can deal with the large-scale DTP (i.e., more than 400 data collection points) efficiently. Moreover, the DTA only uses 5 iterations to obtain the near-optimal solution whereas the normal Genetic Algorithm needs nearly 10000 iterations and still fails to obtain an acceptable solution.

This paper considers an unmanned aerial vehicle based mobile edge computing (UAV based MEC) system, where we assume there is one UAV, acts as an edge cloud, providing data processing services to the Internet of things devices (IoTDs). We consider the UAV hovers at difference places for different time to receive and process data for IoTDs. We aim to minimize the energy consumption of the UAV, including its hovering energy and computation energy, by optimizing the hovering time, scheduling and resource allocation of the tasks received from IoTDs, subject to the quality of service (QoS) requirement of all the IoTDs and the computing resource available at UAV. This is formulated as a mixed-integer non-convex optimization problem, which is difficult to solve in general. We propose an efficient iterative algorithm to get a high-quality suboptimal solution. Simulation results show that our proposed method has a very good performance compared with the other benchmarks.