Modulation of Tropical Cyclone Activity Over the Northwestern Pacific Through the Quasi-Biweekly Oscillation

The quasi-biweekly oscillation(QBWO)is the second most dominant intraseasonal mode for circulation over the Northwestern Pacific(WNP)during boreal summer.In this study,we investigated how the QBWO modulates tropical cyclone(TC)activities over the WNP from dynamic and thermodynamic perspectives.The propagation of the QBWO can be divided into four phases through empirical orthogonal function analysis of the vorticity at 850 hPa,which was proven to be effective in extracting the QBWO signal.TC generation and landings are significantly enhanced during the active period(phases 1 and 2)relative to the inactive period(phases 3 and 4).Composite analyses show the QBWO could significantly modulate TC activity as it propagates northwestward by changing the atmospheric circulation at both high and low levels.Cumulus convection provides an important link between TCs and the QBWO.The major component of the atmosphere heat source is found to be the latent heat release of convection.The condensation latent heat centers,vertical circulation,and water vapor flux divergence cooperate well during different phases of the QBWO.The vertical profile of the condensation latent heat indicates upper-level heating(cooling)during the active(inactive)phases of the QBWO.Thus,the northwestward propagation of the QBWO can modulate TC activity by affecting the configuration of atmospheric heating over the WNP.

基于预算时变的多臂赌博机模型

目前已有很多有关预算的多臂赌博机模型,但这些模型能解决的实际问题具有局限性,即这些问题必须都是全程受一个总预算限制。对此,文中提出基于预算时变的多臂赌博机模型,该模型能够打破这种局限性,并被用于解决其他更多的实际问题。该模型抓住了学习者每一轮的动作都受到相应这一轮预算限制的情况。更具体地说,每一轮,玩家都需要在相应这一轮预算的限制下选择拉L(L≥1)个臂(L不是一个固定值)。玩家的目标就是在每一轮预算的限制下,最大化总的平均奖励。根据这个模型,文中提出基于置信界的动态规划算法。该算法利用模型的特点,每一轮都以臂的经验平均奖励的置信上界为依据,然后使用动态规划算法进行拉臂操作。文中进一步引入遗憾的概念,并从理论上推导得出该算法遗憾的上界与最终预算的总和存在一定的关系。最后,通过实验,将所提算法在不同场景下和其他几个传统的预算受限的多臂赌博机算法(ε-first,KUBE,BTS)进行比较,验证了所提算法的可行性。