Heart disease is a leading global cause of death, while federated learning (FL) is an effective way to predict it. Due to patient privacy concerns and the centralized nature of current FL approaches, collaborative research in heart disease prediction (HDP) faces significant hurdles in computational health systems. This article introduces a distributed online aggregation method within fully decentralized federated learning (DFL), named DeFedHDP, to address these privacy challenges and improve the model of HDP. Moreover, the differential privacy (DP) mechanism is applied to the aggregation strategy of DeFedHDP to protect the privacy of the patients. The data holder communicates directly with neighbors in a series of time-varying directed graphs without the involvement of a central server. Furthermore, each participant is both a trainer of the local model and a collaborator of the other participants’ models. The data does not leave the local device, only the model parameters are exchanged and integrated, and this decentralized approach can further improve the level of privacy protection. In addition, to cope with model gradient disappearance and gradient explosion, the one-point bandit feedback (OPBF) strategy is utilized to estimate the true gradient values. Experiments on a public medical dataset show that the effectiveness of DeFedHDP is close to the centralized FedAVG algorithm for client-server architectures in terms of accuracy and speed.

In Healthcare Internet of Things Networks (HIoTNs), safeguarding the sensitivity of patients’ Electric Healthcare Records (EHRs) is imperative, necessitating effective privacy protection while ensuring ample data for training. Decentralized Federated Learning (DFL), as a peer-to-peer distributed machine learning architecture, serves a pivotal role in preserving the data privacy of EHRs. However, two primary challenges in applying DFL to HIoTNs include safeguarding the privacy of EHRs at individual hospitals and optimizing communication resources among hospitals. This paper proposes a three-layer healthcare framework comprising a physical layer, a communication layer and an edge-encrypted layer to segregate hospital local data from interaction models. Moreover, a novel doubly accelerated DFL algorithm (AccDFL) is introduced to integrate DFL training and information interaction into the communication layer of HIoTNs. By employing the heavy-ball and Nesterov methods, AccDFL achieves double acceleration, ensuring both i-Differential Privacy (DP) and linear convergence. The theoretical analysis of AccDFL delves into the interaction among Distributed Gradient Tracking (DGT), DP, and accelerated mechanisms, presenting comprehensive convergence and privacy analyses to overcome the exponential increase in parameters brought by the acceleration and DP mechanisms. Experimental results with realistic non-IID grayscale and color medical datasets of different disease types affirm the significant advantages of AccDFL over other algorithms in terms of accuracy and communication efficiency.

In this paper, a special class of distributed stochastic nonconvex optimization problem is investigated. Each agent in the network only has access to a local stochastic weakly convex objective function and can only communicate with its neighbours in a random time-delay environment. For solving the considered problem with the distributed communication under time-delay interference, the distributed weakly convex delay-tolerance algorithm (DWDTA) with diminishing step-size and fixed step-size are proposed, respectively. Specifically, we show the convergence of the DWDTA with diminishing step-size by using Moreau Envelope measurement and demonstrate the linear convergence of the DWDTA with fixed step-size under sharpness condition. Our convergence results explicitly characterize the influences of the weakly convex function and the random time-delay interference on the convergence performances, respectively. Finally, numerical results are worked out to verify the effectiveness of the DWDTA.

This paper investigates a class of constrained distributed zeroth-order optimization (ZOO) problems over time-varying unbalanced graphs while ensuring privacy preservation among individual agents. Not taking into account recent progress and addressing these concerns separately, there remains a lack of solutions offering theoretical guarantees for both privacy protection and constrained ZOO over time-varying unbalanced graphs. We hereby propose a novel algorithm, termed the differential privacy (DP) distributed push-sum based zeroth-order constrained optimization algorithm (DP-ZOCOA). Operating over time-varying unbalanced graphs, DP-ZOCOA obviates the need for supplemental suboptimization problem computations, thereby reducing overhead in comparison to distributed primary-dual methods. DP-ZOCOA is specifically tailored to tackle constrained ZOO problems over time-varying unbalanced graphs, offering a guarantee of convergence to the optimal solution while robustly preserving privacy. Moreover, we provide rigorous proofs of convergence and privacy for DP-ZOCOA, underscoring its efficacy in attaining optimal convergence without constraints. To enhance its applicability, we incorporate DP-ZOCOA into the federated learning framework and formulate a decentralized zeroth-order constrained federated learning algorithm (ZOCOA-FL) to address challenges stemming from the time-varying imbalance of communication topology. Finally, the performance and effectiveness of the proposed algorithms are thoroughly evaluated through simulations on distributed least squares (DLS) and decentralized federated learning (DFL) tasks.

We consider the constrained multi-agent online optimization problem in dynamic environments that are vulnerable to Byzantine attacks, where some infiltrated agents may deviate from the prescribed update rule and send arbitrary messages. The objective functions are exposed in a bandit form, i.e., only the function value is revealed to each agent at the sampling instance, and held privately by each agent. The agents only exchange information with their neighbors to update decisions, and the collective goal is to minimize the sum of the unattacked agents' objective functions in dynamic environments, where the same function can only be sampled once. To handle this problem, a Byzantine-Resilient Distributed Bandit Online Convex Optimization (BR-DBOCO) algorithm that can tolerate up to $\mathcal {B}$ Byzantine agents is developed. Specifically, the BR-DBOCO employs the one-point bandit feedback (OPBF) mechanism and state filter to cope with the objective function, which cannot be explicitly expressed in dynamic environments and the arbitrary deviation states caused by Byzantine attacks, respectively. We show that sublinear expected regret is achieved if the accumulative deviation of the comparator sequence also grows sublinearly with a proper exploration parameter. Finally, experimental results are presented to illustrate the effectiveness of the proposed algorithm.