Active learning in semi-supervised classification involves introducing additional labels for unlabelled data to improve the accuracy of the underlying classifier.A challenge is to identify which points to label to bes...Active learning in semi-supervised classification involves introducing additional labels for unlabelled data to improve the accuracy of the underlying classifier.A challenge is to identify which points to label to best improve performance while limiting the number of new labels."Model Change"active learning quantifies the resulting change incurred in the classifier by introducing the additional label(s).We pair this idea with graph-based semi-supervised learning(SSL)methods,that use the spectrum of the graph Laplacian matrix,which can be truncated to avoid prohibitively large computational and storage costs.We consider a family of convex loss functions for which the acquisition function can be efficiently approximated using the Laplace approximation of the posterior distribution.We show a variety of multiclass examples that illustrate improved performance over prior state-of-art.展开更多
This paper introduces Certis, a powerful framework that addresses the challenges of cloud asset tracking, management, and threat detection in modern cybersecurity landscapes. It enhances asset identification and anoma...This paper introduces Certis, a powerful framework that addresses the challenges of cloud asset tracking, management, and threat detection in modern cybersecurity landscapes. It enhances asset identification and anomaly detection through SSL certificate parsing, cloud service provider integration, and advanced fingerprinting techniques like JARM at the application layer. Current work will focus on cross-layer malicious behavior identification to further enhance its capabilities, including minimizing false positives through AI-based learning techniques. Certis promises to offer a powerful solution for organizations seeking proactive cybersecurity defenses in the face of evolving threats.展开更多
Secure Sockets Layer(SSL)and Transport Layer Security(TLS)protocols facilitates a secure framework for identity authentication,data encryption,and message integrity verification.However,with the recent development in ...Secure Sockets Layer(SSL)and Transport Layer Security(TLS)protocols facilitates a secure framework for identity authentication,data encryption,and message integrity verification.However,with the recent development in quantum computing technology,the security of conventional key-based SSL/TLS protocols faces vulnerabilities.In this paper,we propose a scheme by integrating the quantum key into the SSL/TLS framework.Furthermore,the application of post-quantum algorithms is used to enhance and complement the existing encryption suites.Experimental results show that the proposed SSL/TLS communication system based on quantum keys exhibits high performance in latency and throughput.Moreover,the proposed system showcases good resilience against quantum attacks.展开更多
The development of new technologies in smart cities is often hailed as it becomes a necessity to solve many problems like energy consumption and transportation. Wireless networks are part of these technologies but imp...The development of new technologies in smart cities is often hailed as it becomes a necessity to solve many problems like energy consumption and transportation. Wireless networks are part of these technologies but implementation of several antennas, using different frequency bandwidths for many applications might introduce a negative effect on human health security. In wireless networks, most antennas generate sidelobes SSL. SSL causes interference and can be an additional resource for RF power that can affect human being health. This paper aims to study algorithms that can reduce SSL. The study concerns typical uniform linear antenna arrays. Different optimum side lobe level reduction algorithms are presented. Genetic algorithm GA, Chebyshev, and Particle Swarm Optimization algorithm are used in the optimization process. A comparative study between the indicated algorithms in terms of stability, precision, and running time is shown. Results show that using these algorithms in optimizing antenna parameters can reduce SSL. A comparison of these algorithms is carried out and results show the difference between them in terms of running time and SSL reduction Level.展开更多
基金supported by the DOD National Defense Science and Engineering Graduate(NDSEG)Research Fellowshipsupported by the NGA under Contract No.HM04762110003.
文摘Active learning in semi-supervised classification involves introducing additional labels for unlabelled data to improve the accuracy of the underlying classifier.A challenge is to identify which points to label to best improve performance while limiting the number of new labels."Model Change"active learning quantifies the resulting change incurred in the classifier by introducing the additional label(s).We pair this idea with graph-based semi-supervised learning(SSL)methods,that use the spectrum of the graph Laplacian matrix,which can be truncated to avoid prohibitively large computational and storage costs.We consider a family of convex loss functions for which the acquisition function can be efficiently approximated using the Laplace approximation of the posterior distribution.We show a variety of multiclass examples that illustrate improved performance over prior state-of-art.
文摘This paper introduces Certis, a powerful framework that addresses the challenges of cloud asset tracking, management, and threat detection in modern cybersecurity landscapes. It enhances asset identification and anomaly detection through SSL certificate parsing, cloud service provider integration, and advanced fingerprinting techniques like JARM at the application layer. Current work will focus on cross-layer malicious behavior identification to further enhance its capabilities, including minimizing false positives through AI-based learning techniques. Certis promises to offer a powerful solution for organizations seeking proactive cybersecurity defenses in the face of evolving threats.
基金supported by ZTE IndustryUniversityInstitute Cooperation Funds under Grant No.HCCN20221029003.
文摘Secure Sockets Layer(SSL)and Transport Layer Security(TLS)protocols facilitates a secure framework for identity authentication,data encryption,and message integrity verification.However,with the recent development in quantum computing technology,the security of conventional key-based SSL/TLS protocols faces vulnerabilities.In this paper,we propose a scheme by integrating the quantum key into the SSL/TLS framework.Furthermore,the application of post-quantum algorithms is used to enhance and complement the existing encryption suites.Experimental results show that the proposed SSL/TLS communication system based on quantum keys exhibits high performance in latency and throughput.Moreover,the proposed system showcases good resilience against quantum attacks.
文摘The development of new technologies in smart cities is often hailed as it becomes a necessity to solve many problems like energy consumption and transportation. Wireless networks are part of these technologies but implementation of several antennas, using different frequency bandwidths for many applications might introduce a negative effect on human health security. In wireless networks, most antennas generate sidelobes SSL. SSL causes interference and can be an additional resource for RF power that can affect human being health. This paper aims to study algorithms that can reduce SSL. The study concerns typical uniform linear antenna arrays. Different optimum side lobe level reduction algorithms are presented. Genetic algorithm GA, Chebyshev, and Particle Swarm Optimization algorithm are used in the optimization process. A comparative study between the indicated algorithms in terms of stability, precision, and running time is shown. Results show that using these algorithms in optimizing antenna parameters can reduce SSL. A comparison of these algorithms is carried out and results show the difference between them in terms of running time and SSL reduction Level.
