Federated Learning (FL) systems are increasingly deployed across multiple clients to efficiently train a shared model over local data, thereby effectively addressing data silos and reducing communication. However, FL systems are known to be susceptible to model poisoning attacks by malicious clients, who aim at deteriorating the global model accuracy through sending corrupted updates to the central server. Meanwhile, the local accuracy discrepancy among clients, called as performance fairness, could also be exacerbated, which is one of the major concerns of trustworthy FL systems. This paper proposes a novel attack framework called Durable Model Poisoning Attack (DMPA), targeting both fairness and robustness of efficient FL systems. To implement DMPA, we design the over-unlearning strategy, enabling the adversary to generate poisoned updates to compromise partial clients' performance. Furthermore, we develop a dual projection mechanism to improve the durability of model poisoning attacks. Extensive experiments demonstrate that DMPA is powerful and effective even against robust aggregation rules. Particularly, DMPA achieves average$7.6\times$higher reduction of accuracy while decreasing the performance fairness by$3.0\times$compared with baselines. The experiments also indicated that DMPA extends the durability of attack impacts over baselines by$8.5\times$. In addition, experiments in efficient FL systems disclose their vulnerability.

Abstract Due to the inadequate utilization of data correlation and complementarity in the feature extraction process of multimodal remote sensing images, the paper proposes a deep learning semantic segmentation algorithm based on the Dual Channel Attention Mechanism (DCAM). This algorithm uses U‐Net as the backbone, combining the RGB remote sensing image as one input channel with the Convolutional Block Attention Module to extract colour space features. Simultaneously, it utilizes near‐infrared (NIR) as another input channel with the Self‐Attention Module (SAM) to extract shape space features. Finally, by concatenating the multi‐scale attention features of the RGB remote sensing image channel and the NIR remote sensing image channel, it achieves the correlation and complementarity of contextual features between the two modal remote sensing images. Experimental results on the GID‐15 dataset demonstrate that the DCAM algorithm significantly improves the segmentation accuracy, edge segmentation quality, and object segmentation integrity for various types of targets compared to current mainstream segmentation methods.

In a digital library, book search is one of the most important information services. However, with the rapid development of network technologies such as cloud computing, the server‐side of a digital library is becoming more and more untrusted; thus, how to prevent the disclosure of users' book query privacy is causing people's increasingly extensive concern. In this article, we propose to construct a group of plausible fake queries for each user book query to cover up the sensitive subjects behind users' queries. First, we propose a basic framework for the privacy protection in book search, which requires no change to the book search algorithm running on the server‐side, and no compromise to the accuracy of book search. Second, we present a privacy protection model for book search to formulate the constraints that ideal fake queries should satisfy, that is, (i) the feature similarity, which measures the confusion effect of fake queries on users' queries, and (ii) the privacy exposure, which measures the cover‐up effect of fake queries on users' sensitive subjects. Third, we discuss the algorithm implementation for the privacy model. Finally, the effectiveness of our approach is demonstrated by theoretical analysis and experimental evaluation.

BACKGROUND: It is of great clinical significance to develop an accurate computer aided system to accurately diagnose the breast cancer. In this study, an enhanced machine learning framework is established to diagnose the breast cancer. The core of this framework is to adopt fruit fly optimization algorithm (FOA) enhanced by Levy flight (LF) strategy (LFOA) to optimize two key parameters of support vector machine (SVM) and build LFOA-based SVM (LFOA-SVM) for diagnosing the breast cancer. The high-level features abstracted from the volunteers are utilized to diagnose the breast cancer for the first time. RESULTS: In order to verify the effectiveness of the proposed method, 10-fold cross-validation method is used to make comparison among the proposed method, FOA-SVM (model based on original FOA), PSO-SVM (model based on original particle swarm optimization), GA-SVM (model based on genetic algorithm), random forest, back propagation neural network and SVM. The main novelty of LFOA-SVM lies in the combination of FOA with LF strategy that enhances the quality for FOA, thus improving the convergence rate of the FOA optimization process as well as the probability of escaping from local optimal solution. CONCLUSIONS: The experimental results demonstrate that the proposed LFOA-SVM method can beat other counterparts in terms of various performance metrics. It can very well distinguish malignant breast cancer from benign ones and assist the doctor with clinical diagnosis.