The Experts below are selected from a list of 43827 Experts worldwide ranked by ideXlab platform
Liang Xiao - One of the best experts on this subject based on the ideXlab platform.
-
Reinforcement Learning Based Mobile Offloading for Cloud-Based Malware Detection
2017 IEEE Global Communications Conference GLOBECOM 2017 - Proceedings, 2018Co-Authors: Xiaoyue Wan, Geyi Sheng, Liang XiaoAbstract:Cloud-based malware detection improves the detection performance for mobile devices that offload their malware detection tasks to Security Servers with much larger malware database and powerful computational resources. In this paper, we investigate the competition of the radio transmission bandwidths and the data sharing of the Security Server in the dynamic malware detection game, in which each mobile device chooses its offloading rate of the application traces to the Security Server. As the Q-learning technique has a slow learning rate in the game with high dimension, we have designed a mobile malware detection based on hotbooting-Q techniques, which initiates the quality values based on the malware detection experience. We propose an offloading strategy based on deep Q-network technique with a deep convolutional neural network to further improve the detection speed, the detection accuracy, and the utility. Preliminary simulation results verify the detection gain of the scheme compared with the Q-learning based strategy.
-
Cloud-based malware detection game for mobile devices with offloading
IEEE Transactions on Mobile Computing, 2017Co-Authors: Liang Xiao, Xueli HuangAbstract:As accurate malware detection on mobile devices requires fast process of a large number of application traces, cloud-based malware detection can utilize the data sharing and powerful computational resources of Security Servers to improve the detection performance. In this paper, we investigate the cloud-based malware detection game, in which mobile devices offload their application traces to Security Servers via base stations or access points in dynamic networks. We derive the Nash equilibrium (NE) of the static malware detection game and present the existence condition of the NE, showing how mobile devices share their application traces at the Security Server to improve the detection accuracy, and compete for the limited radio bandwidth, the computational and communication resources of the Server. We design a malware detection scheme with Q-learning for a mobile device to derive the optimal offloading rate without knowing the trace generation and the radio bandwidth model of other mobile devices. The detection performance is further improved with the Dyna architecture, in which a mobile device learns from the hypothetical experience to increase its convergence rate. We also design a post-decision state learning-based scheme that utilizes the known radio channel model to accelerate the reinforcement learning process in the malware detection. Simulation results show that the proposed schemes improve the detection accuracy, reduce the detection delay, and increase the utility of a mobile device in the dynamic malware detection game, compared with the benchmark strategy.
-
mobile cloud offloading for malware detections with learning
Conference on Computer Communications Workshops, 2015Co-Authors: Jinliang Liu, Liang XiaoAbstract:Accurate malware detections on mobile devices such as smartphones require fast processing of a large number of data and thus cloud offloading can be used to improve the Security performance of mobile devices with limited resources. The performance of malware detection with cloud offloading depends on the computation speed of the cloud, the population sharing the cloud resources and the bandwidth of the radio access. In the paper, we investigate the offloading rates of smartphones connecting to the same Security Server in a cloud under dynamic network bandwidths and formulate their interactions as a non-cooperative mobile cloud offloading game. The Nash equilibrium of the mobile cloud offloading game and the existence condition are presented. An offloading algorithm based on Q-learning is proposed for smartphones to determine their offloading rates for malware detection with unknown parameters such as transmission costs. Simulation results show that the proposed offloading strategy can achieve the optimal rate and improve the user's utility under dynamic network bandwidths.
Xueli Huang - One of the best experts on this subject based on the ideXlab platform.
-
Cloud-based malware detection game for mobile devices with offloading
IEEE Transactions on Mobile Computing, 2017Co-Authors: Liang Xiao, Xueli HuangAbstract:As accurate malware detection on mobile devices requires fast process of a large number of application traces, cloud-based malware detection can utilize the data sharing and powerful computational resources of Security Servers to improve the detection performance. In this paper, we investigate the cloud-based malware detection game, in which mobile devices offload their application traces to Security Servers via base stations or access points in dynamic networks. We derive the Nash equilibrium (NE) of the static malware detection game and present the existence condition of the NE, showing how mobile devices share their application traces at the Security Server to improve the detection accuracy, and compete for the limited radio bandwidth, the computational and communication resources of the Server. We design a malware detection scheme with Q-learning for a mobile device to derive the optimal offloading rate without knowing the trace generation and the radio bandwidth model of other mobile devices. The detection performance is further improved with the Dyna architecture, in which a mobile device learns from the hypothetical experience to increase its convergence rate. We also design a post-decision state learning-based scheme that utilizes the known radio channel model to accelerate the reinforcement learning process in the malware detection. Simulation results show that the proposed schemes improve the detection accuracy, reduce the detection delay, and increase the utility of a mobile device in the dynamic malware detection game, compared with the benchmark strategy.
Xiaoyue Wan - One of the best experts on this subject based on the ideXlab platform.
-
Reinforcement Learning Based Mobile Offloading for Cloud-Based Malware Detection
2017 IEEE Global Communications Conference GLOBECOM 2017 - Proceedings, 2018Co-Authors: Xiaoyue Wan, Geyi Sheng, Liang XiaoAbstract:Cloud-based malware detection improves the detection performance for mobile devices that offload their malware detection tasks to Security Servers with much larger malware database and powerful computational resources. In this paper, we investigate the competition of the radio transmission bandwidths and the data sharing of the Security Server in the dynamic malware detection game, in which each mobile device chooses its offloading rate of the application traces to the Security Server. As the Q-learning technique has a slow learning rate in the game with high dimension, we have designed a mobile malware detection based on hotbooting-Q techniques, which initiates the quality values based on the malware detection experience. We propose an offloading strategy based on deep Q-network technique with a deep convolutional neural network to further improve the detection speed, the detection accuracy, and the utility. Preliminary simulation results verify the detection gain of the scheme compared with the Q-learning based strategy.
Geyi Sheng - One of the best experts on this subject based on the ideXlab platform.
-
Reinforcement Learning Based Mobile Offloading for Cloud-Based Malware Detection
2017 IEEE Global Communications Conference GLOBECOM 2017 - Proceedings, 2018Co-Authors: Xiaoyue Wan, Geyi Sheng, Liang XiaoAbstract:Cloud-based malware detection improves the detection performance for mobile devices that offload their malware detection tasks to Security Servers with much larger malware database and powerful computational resources. In this paper, we investigate the competition of the radio transmission bandwidths and the data sharing of the Security Server in the dynamic malware detection game, in which each mobile device chooses its offloading rate of the application traces to the Security Server. As the Q-learning technique has a slow learning rate in the game with high dimension, we have designed a mobile malware detection based on hotbooting-Q techniques, which initiates the quality values based on the malware detection experience. We propose an offloading strategy based on deep Q-network technique with a deep convolutional neural network to further improve the detection speed, the detection accuracy, and the utility. Preliminary simulation results verify the detection gain of the scheme compared with the Q-learning based strategy.
Jonathan Guislain - One of the best experts on this subject based on the ideXlab platform.
-
design and validation of a trust based opportunity enabled risk management system
International Conference on Supercomputing, 2017Co-Authors: Alessandro Aldini, Jeanmarc Seigneur, Carlos Ballester Lafuente, Xavier Titi, Jonathan GuislainAbstract:The Bring-Your-Own-Device (BYOD) paradigm favors the use of personal and public devices and communication means in corporate environments, thus representing a challenge for the traditional Security and risk management systems. In this dynamic and heterogeneous setting, the purpose of this paper is to present a methodology called opportunity-enabled risk management (OPPRIM), which supports the decision-making process in access control to remote corporate assets.,OPPRIM relies on a logic-based risk policy model combining estimations of trust, threats and opportunities. Moreover, it is based on a mobile client – Server architecture, where the OPPRIM application running on the user device interacts with the company IT Security Server to manage every access request to corporate assets.,As a mandatory requirement in the highly flexible BYOD setting, in the OPPRIM approach, mobile device Security risks are identified automatically and dynamically depending on the specific environment in which the access request is issued and on the previous history of events.,The main novelty of the OPPRIM approach is the combined treatment of threats (resp., opportunities) and costs (resp., benefits) in a trust-based setting. The OPPRIM system is validated with respect to an economic perspective: cost-benefit sensitivity analysis is conducted through formal methods using the PRISM model checker and through agent-based simulations using the Anylogic framework.