Graph Anomaly Detection (GAD) is a technique used to detect anomalous graph nodes, which is very important in industrial informatics, particularly in cybersecurity, industrial IoT (IIoT), and intelligent monitoring systems. It is widely used for detecting network intrusions, sensor anomalies, and fraudulent behaviors in smart manufacturing, power grids, and industrial automation. In recent years, graph anomaly detection has been mainly based on graph autoencoder (GAE) and comparative learning. Among them, GAE utilizes the reconstruction loss of the graph to estimate the anomalous nodes. However, the existing methods ignore the negative offset problem of abnormal nodes in obtaining node representations. In the graph, there are far more normal nodes than abnormal nodes, and the features of abnormal nodes are diluted by the features of normal nodes during message passing, which affects the model’s ability to detect abnormal nodes. Therefore, we propose AAE-GAD, a graph anomaly detection method based on the attention mechanism of online updates, to solve this problem. Specifically, first, we obtain the local anomaly scores of all nodes by training a local model, extract a set of normal nodes with high confidence, and compute the hybrid attention weights based on the similarity of the nodes’ central representations with the set of normal nodes and the difference of the anomaly scores with their neighboring nodes. Secondly, we get the node representation by modifying the mechanism of message passing and utilizing the hybrid attention weights in the global model to perform dynamic message passing, and the reconstructed graph can better identify the anomalous nodes. Finally, we continuously update the hybrid attention score using the anomaly score obtained from the global model to make it more discriminative. Extensive experiments on six datasets demonstrate the effectiveness of AAE-GAD, and AAE-GAD achieves good results in anomaly detection for contextual and structural types. In highlight, AAE-GAD achieves significant results on the Enron dataset.

This paper presents a novel multi-mode bionic robotic hand. Its bionic finger (BIF) ingeniously combines a magnetic-silica-gel skin with a rigid skeletal framework and integrates a vacuum suction cup at the fingertip. This design enables the bionic manipulator to execute multiple grasping modes, namely enveloping, parallel, and suction grasping. The proposed BIF emulates the skeletal structure of human fingers and equips the fingertip with suction-based grasping functionality, thus achieving both formal bionics and functional superiority. The overall grasping space range of the bionic manipulator can be determined through the computation of the offset of the steel wire, which corresponds to the bending angles of the three joints of the finger. Furthermore, by discerning the four phases within the bionic manipulator’s object-grasping process, in-depth exploration is carried out regarding the unique data characteristics of the magnetic-tactile sensing unit during the grasping operation. On this basis, an accurate prediction of the grasped object’s diameter is achieved. We constructed an autonomous grasping operation platform by integrating an external depth camera with the robotic arm to assess the fundamental performance of this robotic hand in grasping diverse objects.

This article presents AHCA, an agile design framework for Field Programmable Gate Array (FPGA)-based Hashcat acceleration that automates the generation of optimized register transfer level (RTL) code. Our approach is centered on a proposed automated design method using a parameterized domain-specific template (DST) and a specific hardware operator library. The framework analyzes an algorithm’s graph to extract key hardware operators and their interconnection network. To support diverse user inputs, we introduce an innovative operator matching strategy using subgraph isomorphism, which maps algorithms to our operator library. This matched information, combined with design space exploration (DSE), is used to configure the DST and generate the final RTL code, avoiding redundancy for previously implemented algorithms. Compared to state-of-the-art high-level synthesis (HLS) tools, AHCA demonstrates a maximum performance enhancement of 797×, a Look-Up table (LUT) efficiency improvement of up to 105×, and an energy efficiency gain of up to 676×. When deployed on an FPGA for password cracking, the AHCA-generated hardware achieves a 63.95× enhancement in energy efficiency over CPUs and a 4.71× improvement over GPUs.

This article aims to investigate the issue of energy management for multienergy communities with diversified energy consumer profiles. Since the energy consumers have their points of parity and differences in their demographic, psychological, and behavioral traits, it is assumed that the energy management scheme should be capable of meeting their actual needs instead of just reducing the energy bills as conventional researches do. Hence, this article proposes a multiobjective energy management strategy for diversified energy consumers to achieve energy-tailoring. First, a novel entropy based energy consumer clustering approach is proposed for optimal consumer segmentation. Following that, four multiobjective energy management models are proposed to achieve the goal of energy bill minimization, maximization of green energy usage, reduction of energy losses and optimization of energy usage quality. Meanwhile, the strategy to achieve priority-based coordination of the four objectives is developed. To this end, a multiagent system is developed to perform the optimization model. Simulation case studies are conducted to validate the effectiveness of the proposed method. Numerical results have shown that the proposed approach can achieve co-optimization of the four objectives, and the tradeoffs caused by the competing interest groups are maintained at a low level.