Space-Air-Ground Integrated Mobile Crowdsensing for Partially Observable Data Collection by Multi-Scale Convolutional Graph Reinforcement Learning

Mobile crowdsensing (MCS) is attracting considerable attention in the past few years as a new paradigm for large-scale information sensing. Unmanned aerial vehicles (UAVs) have played a significant role in MCS tasks and served as crucial nodes in the newly-proposed space-air-ground integrated network (SAGIN). In this paper, we incorporate SAGIN into MCS task and present a Space-Air-Ground integrated Mobile CrowdSensing (SAG-MCS) problem. Based on multi-source observations from embedded sensors and satellites, an aerial UAV swarm is required to carry out energy-efficient data collection and recharging tasks. Up to date, few studies have explored such multi-task MCS problem with the cooperation of UAV swarm and satellites. To address this multi-agent problem, we propose a novel deep reinforcement learning (DRL) based method called Multi-Scale Soft Deep Recurrent Graph Network (ms-SDRGN). Our ms-SDRGN approach incorporates a multi-scale convolutional encoder to process multi-source raw observations for better feature exploitation. We also use a graph attention mechanism to model inter-UAV communications and aggregate extra neighboring information, and utilize a gated recurrent unit for long-term performance. In addition, a stochastic policy can be learned through a maximum-entropy method with an adjustable temperature parameter. Specifically, we design a heuristic reward function to encourage the agents to achieve global cooperation under partial observability. We train the model to convergence and conduct a series of case studies. Evaluation results show statistical significance and that ms-SDRGN outperforms three state-of-the-art DRL baselines in SAG-MCS. Compared with the best-performing baseline, ms-SDRGN improves 29.0% reward and 3.8% CFE score. We also investigate the scalability and robustness of ms-SDRGN towards DRL environments with diverse observation scales or demanding communication conditions.     

基于程序依赖图的Aspect挖掘技术

为了有效地识别遗产代码中因横切关注点而导致的零散代码或交织代码,提出了一种基于程序依赖图的Aspect挖掘方法.根据横切关注点的本质特性,考察了横切关注点与核心关注点之间的依赖关系及其在依赖图上的表现形式,对于单个方法体,结合程序的语义信息,在程序依赖图中引入一个虚拟结点表征方法体的核心功能,所有影响核心功能的语句都与之建立依赖关系,然后对依赖图进行拓扑分析,先从虚拟结点开始分离出核心代码,再通过求解连通分量的方式挖掘方法体内横切代码.实例分析表明,该方法不仅能识别出方法体内的零散特性代码,而且还能有效地挖掘出交织特性的代码.     

Halin图的L(d,l)标号

给定图G和正整数d,图G的L(d,1)标号是指从图G的顶点集到非负整数集的一个映射f满足:对任意的x,y∈V(G),当dG(x,y)=1时,有|f(x)-f(y)|≥d;当dG(x,y)=2时,有|f(x)-f(y)|≥1。图G的L(d,1)标号数λd(G)是指最小的正整数k使得G有一个L(d,1)标号f满足f(V){0,1,2,…,k}。已知对于最大度为Δ的一般图有λd(G)≤Δ2 (d-1)Δ。讨论了Halin图的L(d,1)标号问题,证明了λd(G)≤Δ 3(2d-1)。     

Weak Graph Map Homotopy and Its Applications∗

The authors introduce a notion of a weak graph map homotopy (they call it M-homotopy), discuss its properties and applications. They prove that the weak graph map homotopy equivalence between graphs coincides with the graph homotopy equivalence defined by Yau et al in 2001. The difference between them is that the weak graph map homotopy transformation is defined in terms of maps, while the graph homotopy transformation is defined by means of combinatorial operations. They discuss its advantages over the graph homotopy transformation. As its applications, they investigate the mapping class group of a graph and the 1-order MP-homotopy group of a pointed simple graph. Moreover, they show that the 1-order MP-homotopy group of a pointed simple graph is invariant up to the weak graph map homotopy equivalence.     

若干倍图的邻点可区别均匀全染色

研究一些倍图的邻点可区别均匀全染色(AVDETC), 利用构造法和匹配法给出了偶阶完全图、 偶阶圈、 路、 星和轮的倍图的邻点可区别均匀全色数, 并验证了它们满足邻点可区别均匀全染色猜想(AVDETCC).     

图Cn∪Pm的算术标号

设Ｃｎ∪Ｐｍ（ｎ≥３，ｍ≥２，ｎ，ｍ∈Ｎ）表示一个圈Ｃｎ和一条与其不相交的路Ｐｍ组成的图，本文证明图Ｃｎ∪Ｐｍ是算术图。     

对集不交的循环赛图K11^（i）与对集的算法

给出了边矩阵和循环赛图的定义。提出了求解完全图K（2n＋1）的△（G）＋1个对集最的算法，以及对集互交的循环赛图K11^（1），K11^（2），…，K11^（i）的构造方法。讨论任意对集Ei及循环图K（2n＋1）^＊的个数问题。介绍了14个对集不交的循环赛图K11^（1），K11^（2），…，K11^（14）的构造过程。     

蕴含K1，1，3的正可图序列的最小度和

Gould R J等人考虑了下述经典Turán型极值问题的变形对于给定的图H,确定最小的正偶数σ(H,n),使得对于每一个n项正可图序列π=(d1,d2,...,dn),当σ(π)=d1+d2+...+dn≥σ(H,n)时,π有一个实现G以H作为子图.本文完全确定了σ(K1,1,3,n)之值,其中Kr,s,t是r×s×t完全三部图.     

第二小阶数的点本原1/2-弧传递图

通过研究群PSL(3,3)在二维射影空间PV(2 ,3)中极大独立点集之集合上的本原作用的次轨道结构 ,构造了一个阶数为 2 34,度数为 48,自同构群为Aut(PSL(3,) )的点本原 12 弧传递图 ,从而发现了第二小阶数的点本原 12 弧传递图 .     

基于时空建模的动态图卷积神经网络

为了使图表示学习得到的嵌入向量对节点和边不断变化的动态图具有很好的信息表征能力,提出一种动态图卷积神经网络模型(DyGCN),将动态图上的表示学习建模为时间和空间信息的聚合.该模型将从图卷积神经网络(GCN)的空间卷积提取图上的结构信息与从时间卷积神经网络(TCN)的因果卷积提取时序上的历史信息相结合,同时在空间卷积层...