Zero-shot learning (ZSL) tackles the novel class recognition problem by transferring semantic knowledge from seen classes to unseen ones. Semantic knowledge is typically represented by attribute descriptions shared between different classes, which act as strong priors for localizing object attributes that represent discriminative region features, enabling significant and sufficient visual-semantic interaction for advancing ZSL. Existing attention-based models have struggled to learn inferior region features in a single image by solely using unidirectional attention, which ignore the transferable and discriminative attribute localization of visual features for representing the key semantic knowledge for effective knowledge transfer in ZSL. In this paper, we propose a cross attribute-guided Transformer network, termed TransZero++, to refine visual features and learn accurate attribute localization for key semantic knowledge representations in ZSL. Specifically, TransZero++ employs an attribute $\rightarrow$ visual Transformer sub-net (AVT) and a visual $\rightarrow$ attribute Transformer sub-net (VAT) to learn attribute-based visual features and visual-based attribute features, respectively. By further introducing feature-level and prediction-level semantical collaborative losses, the two attribute-guided transformers teach each other to learn semantic-augmented visual embeddings for key semantic knowledge representations via semantical collaborative learning. Finally, the semantic-augmented visual embeddings learned by AVT and VAT are fused to conduct desirable visual-semantic interaction cooperated with class semantic vectors for ZSL classification. Extensive experiments show that TransZero++ achieves the new state-of-the-art results on three golden ZSL benchmarks and on the large-scale ImageNet dataset. The project website is available at: https://shiming-chen.github.io/TransZero-pp/TransZero-pp.html .

In this article, a hierarchical task offloading strategy is presented for delay-tolerant and delay-sensitive missions by integrating edge computing and artificial intelligence into Cybertwin-based network to guarantee user Quality of Experience (QoE), low latency, and ultrareliable services, which are huge challenges to the Internet of Things (IoT) due to diverse application requirements, heterogeneous multidimensional resources, and time-varying network environments. The novel scheme achieves faster task processing, dynamic real-time allocation, and lower overhead by taking advantages of a multiagent deep deterministic policy gradient (MADDPG). Moreover, federated learning is used to train the MADDPG model. Numerical results demonstrate that the proposed algorithm improves system processing efficiency and task completion ratio compared to the benchmark schemes.

The introduction of information and communication technologies makes network environments increasingly open, leaving smart-grid control systems incredibly vulnerable to malicious attacks. False data injection (FDI) attacks stealthily tamper with measurement data, resulting in erroneous decisions made by the control center that greatly influence the normal operation of the power system. By taking advantage of real-time data acquisition with edge computing, in this article, we propose a scheme based on classification of predicted residuals (CPRs) for the FDI attack detection. The CPR scheme first predicts the acquired measurement data at the edge of the sensing network via developing an accurate prediction model. Followed the novel real-time classification method under the edge devices supporting, it classifies the predicted residuals independent of the false data to enhance the detection accuracy. Through these two steps, the detection rate of FDI attacks is greatly improved. The proposed scheme is validated in a real microgrid testbed. Experimental results show that the CPR scheme performs well in detecting FDI attacks and remains sensitive in injection attack probability and magnitude. The detection scheme even has effectiveness at low injection attack probability and magnitude (5% and 0.018 per thousand, respectively). Furthermore, it also proves that the proposed scheme has applicability in high real-time requirements at the edge of smart grids.

Advanced communication and information technologies enable smart grids to be more intelligent and automated, although many security issues are emerging. Security situational awareness (SSA) has been envisioned as a potential approach to provide safe services for power systems’ operation. However, in the power cloud master station mode, massive heterogeneous power terminals make SSA complicated, and failure information cannot be promptly delivered. Moreover, the dynamic and continuous situational space also increases the challenges of SSA. By taking advantages of edge intelligence, this paper introduces edge computing between terminals and the cloud to address the drawbacks of the traditional power cloud paradigm. Moreover, a deep reinforcement learning algorithm based on the edge computing paradigm of multiagent deep deterministic policy gradient (MADDPG) is proposed. The minimum processing cost under the premise of minimum detection error rate is taken to analyze the smart grids’ SSA. Performance evaluations show that the algorithm under this paradigm can achieve faster convergence and the optimal goal, namely the provision of real-time protection for smart grids.

Residing in the proximity of end devices, edge computing (EC) holds great potential to provide low-latency, energy-efficient, and secure services, which has become an essential part of the Industrial Internet of Things (IIoT). To future accelerate task processing and reduce service latency, this work proposes an online incentive-driven task allocation scheme to stimulate collaborative computing among EC servers and IIoT devices. To better serve dynamic and heterogeneous tasks in terms of profiles and importance, EC servers (including neighboring servers) and IIoT devices with available resources can cooperatively process the tasks. Considering the heterogeneity of computing resources in edge servers and industrial IoT devices, we formulate a task allocation problem, which is NP hard. An online incentive-driven task allocation algorithm is proposed to this NP-hard problem, which will optimize task assignment strategies to maximize system utility, promote faster computing, and stimulate collaborative computing. Theoretical analyses show that the online incentive algorithm can satisfy incentive compatibility, individual rationality, computational efficiency, and feasibility. The results demonstrate that the proposed task allocation scheme with collaborative EC achieves superior performance and effectiveness.

Radio frequency fingerprinting (RFF) is a lightweight authentication technology for resource-limited terminal nodes by exploiting the unique hardware imperfections resulting from the manufacturing process. Previous studies about radio frequency fingerprinting identification (RFFI) mainly concentrate on improving the accuracy which is evaluated by the single receiver device that trains and identifies all the nodes. Due to the mobility of the consumer electronic terminals, these terminal nodes may need to be identified by the different receivers. In this paper, we propose a cross-device radio frequency fingerprinting identification scheme which allows enrolled nodes to be authenticated by different devices. Motivated by the observation that signals collected by different receiver devices have a distribution shift that would violate the basic independent and identically distributed (i.i.d) assumption of supervised learning. Domain adaptation is adopted to improve the accuracy under different receivers, which can align the data captured from different devices and eliminate the distribution shift through the labeled data from one receiver device and unlabeled data from the other device. By this way, the distribution shift from different devices is corrected. Extensive experiment configurations under various Signal-to-noise ratio (SNR) are carried out to demonstrate the performance of domain adaptation with the same model structure. The results indicate that classification accuracy under different devices can be increased by 7%-15% and get a stable accuracy rate higher than 90% by leveraging our proposed cross-device RFFI scheme.