The exponentially-distributed random timestepping algorithm with boundary test is implemented to evaluate the prices of some variety of single one-sided barrier option contracts within the framework of Black-Scholes m...The exponentially-distributed random timestepping algorithm with boundary test is implemented to evaluate the prices of some variety of single one-sided barrier option contracts within the framework of Black-Scholes model, giving efficient estimation of their hitting times. It is numerically shown that this algorithm, as for the Brownian bridge technique, can improve the rate of weak convergence from order one-half for the standard Monte Carlo to order 1. The exponential timestepping algorithm, however, displays better results, for a given amount of CPU time, than the Brownian bridge technique as the step size becomes larger or the volatility grows up. This is due to the features of the exponential distribution which is more strongly peaked near the origin and has a higher kurtosis compared to the normal distribution, giving more stability of the exponential timestepping algorithm at large time steps and high levels of volatility.展开更多
为了增加网络吞吐量并改善用户体验,提出一种基于Q学习(Q-learning)的多业务网络选择博弈(Multi-Service Network Selection Game based on Q-learning,QSNG)策略。该策略通过模糊推理和综合属性评估获得多业务网络效用函数,并将其用作Q...为了增加网络吞吐量并改善用户体验,提出一种基于Q学习(Q-learning)的多业务网络选择博弈(Multi-Service Network Selection Game based on Q-learning,QSNG)策略。该策略通过模糊推理和综合属性评估获得多业务网络效用函数,并将其用作Q-learning的奖励。用户通过博弈算法预测网络选择策略收益,避免访问负载较重的网络。同时,使用二进制指数退避算法减少多个用户并发访问某个网络的概率。仿真结果表明,所提策略可以根据用户的QoS需求和价格偏好自适应地切换到最合适的网络,将其与基于强化学习的网络辅助反馈(Reinforcement Learning with Network-Assisted Feedback,RLNF)策略和无线网络选择博弈(Radio Network Selection Games,RSG)策略相比,所提策略可以分别减少总切换数量的80%和60%,使网络吞吐量分别提高了7%和8%,并且可以保证系统的公平性。展开更多
文摘The exponentially-distributed random timestepping algorithm with boundary test is implemented to evaluate the prices of some variety of single one-sided barrier option contracts within the framework of Black-Scholes model, giving efficient estimation of their hitting times. It is numerically shown that this algorithm, as for the Brownian bridge technique, can improve the rate of weak convergence from order one-half for the standard Monte Carlo to order 1. The exponential timestepping algorithm, however, displays better results, for a given amount of CPU time, than the Brownian bridge technique as the step size becomes larger or the volatility grows up. This is due to the features of the exponential distribution which is more strongly peaked near the origin and has a higher kurtosis compared to the normal distribution, giving more stability of the exponential timestepping algorithm at large time steps and high levels of volatility.
文摘为了增加网络吞吐量并改善用户体验,提出一种基于Q学习(Q-learning)的多业务网络选择博弈(Multi-Service Network Selection Game based on Q-learning,QSNG)策略。该策略通过模糊推理和综合属性评估获得多业务网络效用函数,并将其用作Q-learning的奖励。用户通过博弈算法预测网络选择策略收益,避免访问负载较重的网络。同时,使用二进制指数退避算法减少多个用户并发访问某个网络的概率。仿真结果表明,所提策略可以根据用户的QoS需求和价格偏好自适应地切换到最合适的网络,将其与基于强化学习的网络辅助反馈(Reinforcement Learning with Network-Assisted Feedback,RLNF)策略和无线网络选择博弈(Radio Network Selection Games,RSG)策略相比,所提策略可以分别减少总切换数量的80%和60%,使网络吞吐量分别提高了7%和8%,并且可以保证系统的公平性。
