Search

Deep learning‐based models have become ubiquitous across a wide range of applications, including computer vision, natural language processing, and robotics. Despite their efficacy, one of the significant challenges associated with deep neural network (DNN) models is the potential risk of copyright leakage due to the inherent vulnerability of the entire model architecture and the communication burden of the large models during publishing. So far, it is still challenging for us to safeguard the intellectual property rights of these DNN models while reducing the communication time during model publishing. To this end, this paper introduces a novel approach using knowledge distillation techniques aimed at training a surrogate model to stand in for the original DNN model. To be specific, a knowledge distillation generative adversarial network (KDGAN) model is proposed to train a student model capable of achieving remarkable performance levels while simultaneously safeguarding the copyright integrity of the original large teacher model and improving communication efficiency during model publishing. Herein, comprehensive experiments are conducted to showcase the efficacy of model copyright protection, communication‐efficient model publishing, and the superiority of the proposed KDGAN model over other copyright protection mechanisms. [ABSTRACT FROM AUTHOR]

Smart grids fall at the intersection of conventional energy systems and modern informatics in the present digitalized energy environment. The growing number of linked devices and sensors in these networks leads to the generation of complex structures and vast quantities of data, presenting benefits and challenges. Safeguarding these complex structures against malicious intrusions and illegal activities is an important problem. The paper's main objective is to enhance smart grid security by utilizing the data mining and Artificial Intelligence (AI) approaches. As huge amounts of data are collected from the smart grids based on tiny and smart internet of things (IoT) devices, this data poses challenges as well as provides opportunities. The challenges come from analyzing this huge data, especially in real-time. At the same time, it provides opportunities to enhance the smart grid services and protection. Therefore, to overcome these challenges, this paper proposes a feedforward deep learning approach for data mining to secure the smart grid from different anomalies and allow the system to adapt to any risk it might face. Deep learning will allow the system to adjust dynamically to emerging risks. The proposed system has been examined using Power System Attack Datasets sourced from the Mississippi State University and Oak Ridge National Laboratory. The results show a detection accuracy of 91% just using 50% of the dataset features. Different percentages of the features are examined as well. However, we concluded that 50% of the features are enough for identifying the smart grid risks based on the given dataset. [ABSTRACT FROM AUTHOR]