A Service Level Agreement(SLA) is a legal contract between any two parties to ensure an adequate Quality of Service(Qo S). Most research on SLAs has concentrated on protecting the user data through encryption. However...A Service Level Agreement(SLA) is a legal contract between any two parties to ensure an adequate Quality of Service(Qo S). Most research on SLAs has concentrated on protecting the user data through encryption. However, these methods can not supervise a cloud service provider(CSP) directly. In order to address this problem, we propose a privacy-based SLA violation detection model for cloud computing based on Markov decision process theory. This model can recognize and regulate CSP's actions based on specific requirements of various users. Additionally, the model could make effective evaluation to the credibility of CSP, and can monitor events that user privacy is violated. Experiments and analysis indicate that the violation detection model can achieve good results in both the algorithm's convergence and prediction effect.展开更多
One of the main goals of machine learning is to study the generalization performance of learning algorithms. The previous main results describing the generalization ability of learning algorithms are usually based on ...One of the main goals of machine learning is to study the generalization performance of learning algorithms. The previous main results describing the generalization ability of learning algorithms are usually based on independent and identically distributed （i.i.d.） samples. However, independence is a very restrictive concept for both theory and real-world applications. In this paper we go far beyond this classical framework by establishing the bounds on the rate of relative uniform convergence for the Empirical Risk Minimization （ERM） algorithm with uniformly ergodic Markov chain samples. We not only obtain generalization bounds of ERM algorithm, but also show that the ERM algorithm with uniformly ergodic Markov chain samples is consistent. The established theory underlies application of ERM type of learning algorithms.展开更多
基金supported in part by National Natural Science Foundation of China (NSFC) under Grant U1509219 and 2017YFB0802900
文摘A Service Level Agreement(SLA) is a legal contract between any two parties to ensure an adequate Quality of Service(Qo S). Most research on SLAs has concentrated on protecting the user data through encryption. However, these methods can not supervise a cloud service provider(CSP) directly. In order to address this problem, we propose a privacy-based SLA violation detection model for cloud computing based on Markov decision process theory. This model can recognize and regulate CSP's actions based on specific requirements of various users. Additionally, the model could make effective evaluation to the credibility of CSP, and can monitor events that user privacy is violated. Experiments and analysis indicate that the violation detection model can achieve good results in both the algorithm's convergence and prediction effect.
基金Supported by National 973 project(No.2013CB329404)Key Project of NSF of China(No.11131006)National Natural Science Foundation of China(No.61075054,61370000)
文摘One of the main goals of machine learning is to study the generalization performance of learning algorithms. The previous main results describing the generalization ability of learning algorithms are usually based on independent and identically distributed （i.i.d.） samples. However, independence is a very restrictive concept for both theory and real-world applications. In this paper we go far beyond this classical framework by establishing the bounds on the rate of relative uniform convergence for the Empirical Risk Minimization （ERM） algorithm with uniformly ergodic Markov chain samples. We not only obtain generalization bounds of ERM algorithm, but also show that the ERM algorithm with uniformly ergodic Markov chain samples is consistent. The established theory underlies application of ERM type of learning algorithms.
