Graphs are playing a crucial role in different fields since they are powerful tools to unveil intrinsic relationships among signals. In many scenarios, an accurate graph structure representing signals is not available at all and that motivates people to learn a reliable graph structure directly from observed signals. However, in real life, it is inevitable that there exists uncertainty in the observed signals due to noise measurements or limited observability, which causes a reduction in reliability of the learned graph. To this end, we propose a graph learning framework using Wasserstein distributionally robust optimization (WDRO) which handles uncertainty in data by defining an uncertainty set on distributions of the observed data. Specifically, two models are developed, one of which assumes all distributions in uncertainty set are Gaussian distributions and the other one has no prior distributional assumption. Instead of using interior point method directly, we propose two algorithms to solve the corresponding models and show that our algorithms are more time-saving. In addition, we also reformulate both two models into Semi-Definite Programming (SDP), and illustrate that they are intractable in the scenario of large-scale graph. Experiments on both synthetic and real world data are carried out to validate the proposed framework, which show that our scheme can learn a reliable graph in the context of uncertainty.
Purpose Pervasive computing enhances the environment by embedding many computers that are gracefully integrated with human users. The purpose of this paper is to describe the creation of a smart context‐aware environment in which computation follows people and serves them everywhere. Building such smart environments is still difficult and complex due to lacking a uniform infrastructure that can adapt to diverse smart domains. Design/methodology/approach To address this problem, the paper proposes an agent‐based pluggable infrastructure which integrates a mobile agent system named pvMogent, establishes an ontology‐based context model and introduces a workflow‐based application model with the open services gateway initiative (OSGi) framework. By plugging corresponding domain context in ontology model and different applications, the infrastructure can be customized to various domains. Findings Through the implementation of several context‐aware applications, it was found that the infrastructure can largely reduce the development complexity as well as keep the domain extensibility by plugging corresponding domain context in ontology model. Originality/value In this paper, a number of key techniques are explored which are suitable for building context‐awareness. The experiences and lessons learned from the system development could further facilitate and inspire the research in this direction.
This chapter focuses on the feature generation problem for graphs and networks. It discusses the feature types. Based on the scope where the features are computed, existing features can be divided into neighborhood-level features and global-level features. The chapter describes the existing feature generation methods and divide them into feature extraction approaches and feature learning approaches. It presents several applications to illustrate feature usages. The chapter discusses the applications of multi-label classification, link prediction, anomaly detection, and visualization. It focuses on the neighborhood-level features as well as their applications in graph analysis tasks. The chapter describes a representative factorization-based method, and discusses its differences from the neural network methods. Different from multi-label prediction, link prediction involves pairs of nodes instead of individual nodes.
This paper presents an Object-Oriented class library for scanning path generation in SLS/SLM (Selective Laser Sintering/Selective Laser Melting) process. The classes in the library meet the minimal requirement for the scanning path generation. Specially, in order to take advantage of the Multiprocessor technology and save the generation time, parallel computing is considered in the class library. At last, an application was developed using the class library and an experiment is provided to verify the feasibility of the parallel computing algorithm in the library.
Neuro-symbolic systems combine the abilities of neural perception and logical reasoning. However, end-to-end learning of neuro-symbolic systems is still an unsolved challenge. This paper proposes a natural framework that fuses neural network training, symbol grounding, and logical constraint synthesis into a coherent and efficient end-to-end learning process. The capability of this framework comes from the improved interactions between the neural and the symbolic parts of the system in both the training and inference stages. Technically, to bridge the gap between the continuous neural network and the discrete logical constraint, we introduce a difference-of-convex programming technique to relax the logical constraints while maintaining their precision. We also employ cardinality constraints as the language for logical constraint learning and incorporate a trust region method to avoid the degeneracy of logical constraint in learning. Both theoretical analyses and empirical evaluations substantiate the effectiveness of the proposed framework.
This paper establishes an AIDS model based on the spread characteristics in China, spread from high-risk to general population, through bridge population based on cellular automata.Matlab is used to simulate the dynamic performance of the model, the simulated results are consistent with statistics from Estimates for the HIV/AIDS Epidemic in China, which shows the feasibility and effectiveness of the proposed model.Then, a simulation of the effects of bridge population on AIDS spread is done and effective optimized control strategies are offered.
In the object-oriented dynamic programming environment, dynamic modification of a class, which permits change of it at run-time and without recompilation, is the key point to exploit flexibility and support rapid prototyping. However, it causes a problem that the existing objects of the modified class are difficult to handle. In this paper, the concept of "cloned class" is introduced, and a method based on it for modifying dynamically classes is proposed.
上下文相关图文法是描述可视化语言的形式化工具.为了直观地刻画并高效地分析可视化语言,已有图文法形式框架均着重于文法形式和分析算法的研究,而忽略了对它们之间表达能力的分析.在对已有上下文相关图文法形式框架的关键特征进行分析和归纳的基础上,通过构造不同形式框架之间的转换算法,揭示并形式化证明了它们表达能力之间的关系.而且,转换算法在不同形式框架之间建立了关联,使图文法的应用不必再局限于一个框架,而是可以选择不同框架分别进行图的描述和分析,从而提高了上下文相关图文法的易用性.;Context-Sensitive graph grammars are formal tools used for specifying visual languages. In order to intuitively describe and parse visual languages, current research has stressed the formalisms and algorithms of graph grammars, but has neglected the comparison of their expressiveness. Based on the analysis and induction of the key characteristics of context-sensitive graph grammar, the relationships between their expressiveness are uncovered and proved in this paper by constructing formalism-transforming algorithms. Moreover, the proposed algorithms correlate with these formalisms; thus, facilitating the usage of context-sensitive graph grammars, as alternative formalisms rather than merely one can be chosen to separately specify and parse visual objects in applications.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)
'/%3e%3cuse xlink:href='%23a' transform='rotate(144 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(216 450 300)'/%3e%3cuse xlink:href='%23a' transform='rotate(288 450 300)'/%3e%3c/svg%3e)
'/%3e%3c/defs%3e%3cpath fill='%23012169' d='M0 0h10080v5040H0z'/%3e%3cpath stroke='%23fff' stroke-width='.6' d='m0 0 6 3m0-3L0 3' clip-path='url(%23b)' transform='scale(840)'/%3e%3cpath stroke='%23e4002b' stroke-width='.4' d='m0 0 6 3m0-3L0 3' clip-path='url(%23c)' transform='scale(840)'/%3e%3cpath stroke='%23fff' stroke-width='840' d='M2520 0v2520M0 1260h5040'/%3e%3cpath stroke='%23e4002b' stroke-width='504' d='M2520 0v2520M0 1260h5040'/%3e%3cg fill='%23fff'%3e%3cuse xlink:href='%23d' x='2520' y='3780'/%3e%3cuse xlink:href='%23a' x='7560' y='4200'/%3e%3cuse xlink:href='%23a' x='6300' y='2205'/%3e%3cuse xlink:href='%23a' x='7560' y='840'/%3e%3cuse xlink:href='%23a' x='8680' y='1869'/%3e%3cuse xlink:href='%23e' x='8064' y='2730'/%3e%3c/g%3e%3c/svg%3e)
