Learning Bayesian network structure with immune algorithm

Finding out reasonable structures from bulky data is one of the difficulties in modeling of Bayesian network （BN）, which is also necessary in promoting the application of BN. This pa- per proposes an immune algorithm based method （BN-IA） for the learning of the BN structure with the idea of vaccination. Further- more, the methods on how to extract the effective vaccines from local optimal structure and root nodes are also described in details. Finally, the simulation studies are implemented with the helicopter convertor BN model and the car start BN model. The comparison results show that the proposed vaccines and the BN-IA can learn the BN structure effectively and efficiently.

Integrated importance measure for multi-state coherent systems of k level

To verify the effectiveness of the integrated importance measure （IIM） for multi-state coherent systems of k level, the definition and physical meaning of IIM are demonstrated. Then, the improvement potential and Δ-importance measures are generalized to multi-state coherent systems based on the system performance level, and the relationships between IIM and traditional importance measures are discussed. The characteristics of IIM are demonstrated in both series and parallel systems. Also, an application to an oil transportation system is given. The comparison results show that： （i） IIM has some useful properties that are not possessed by traditional importance measures; （ii） IIM is effective in evaluating the component role in multi-state systems when the component reliability and the failure rate are simultaneously considered.

Performance Measures for Systems Under Multiple Environments

In this paper, the system which operates in multiple environments is studied. The environment process is governed by a Markov process, and the deterioration process is governed by another Markov process given the system in a certain environment. In terms of the above two processes, a new Markov process is constructed to represent the evolution of the system.In terms of Ion-Channel modeling theory, Markov process theory and matrix partition method, some reliability indexes for the system are obtained, i.e., system reliability, environment reliability, system multiple-interval reliability, system availability,environment availability, system multiple-point availability, etc.Finally, a numerical example is given to illustrate the results obtained in the paper.

Identification of unstable individuals in dynamic networks

To identify the unstable individuals of networks is of great importance for information mining and security management.Exploring a broad range of steady-state dynamical processes including biochemical dynamics,epidemic processes,birth-death processes and regulatory dynamics,we propose a new index from the microscopic perspective to measure the stability of network nodes based on the local correlation matrix.The proposed index describes the stability of each node based on the activity change of the node after its neighbor is disturbed.Simulation and comparison results show our index can identify the most unstable nodes in the network with various dynamical behaviors,which would actually create a richer way and a novel insight of exploring the problem of network controlling and optimization.

一种基于血液检测的诊断和预测HBV相关疾病的弹性方法

Chronic hepatitis B virus(HBV)infection,which threatens global public health,is a major contributor to liver-related morbidity and mortality.Examinations for liver diseases related to chronic HBV infection-including laboratory tests,ultrasounds,computed tomography(CT),and liver biopsies-may take up medical resources,particularly since they overlap in most instances.Thus,there is an urgent need to establish an economical and effective diagnosis method in order to streamline the medical process for HBV-related disea ses.Using complex network models constructed based on clinical blood tests,we provide such a method by defining the novel measure of functional resilience to assess patients’liver conditions.By combining network models and dynamics,we discovered the pivotal items and their corresponding thresholds,which can guide further research on preventing disease deterioration in critical states of these diseases.The macro-averaged precision of our method,functional resilience,is84.74%,whereas the macro-averaged precision of physicians’experience without assistance from imaging or biopsy is 55.63%.From an economic perspective,our approach could save the equivalent of at least30 USD per visit for most Chinese patients and at least 400 USD per visit for most US patients,compared with general diagnostic methods.Globally,this will add to savings of at least 10.5 billion USD annually.Our method can comprehensively evaluate the condition of patients’livers and help avert the waste of medical resources during the diagnosis of liver disease by reducing excessive imaging exams.

Recent advances in system reliability optimization driven by importance measures

System reliability optimization problems have been widely discussed to maximize system reliability with resource constraints.Bimbaum importance is a wellknown method for evaluating the effect of component reliability on system reliability.Many importance measures(IMs)are extended for binary,multistate,and continuous systems from different aspects based on the Bimbaum importance.Recently,these IMs have been applied in allocating limited resources to the component to maximize system performance.Therefore,the significance of Bimbaum importance is illustrated from the perspective of probability principle and gradient geometrical sense.Furthermore,the equations of various extended IMs are provided subsequently.The rules for simple optimization problems are summarized to enhance system reliability by using ranking or heuristic methods based on IMs.The importance-based optimization algorithms for complex or large-scale systems are generalized to obtain remarkable solutions by using IM-based local search or simplification methods.Furthermore,a general framework driven by IM is developed to solve optimization problems.Finally,some challenges in system reliability optimization that need to be solved in the future are presented.

Novel interpretable mechanism of neural networks based on network decoupling method

The lack of interpretability of the neural network algorithm has become the bottleneck of its wide application.We propose a general mathematical framework,which couples the complex structure of the system with the nonlinear activation function to explore the decoupled dimension reduction method of high-dime sional system and reveal the calculation mechanism of the neural network.We apply our framework to some network models and a real system of the whole neuron map of Caenorhabditis elegans.Result shows that a simple linear mapping relationship exists between network structure and network behavior in the neural network with highdimensional and nonlinear characteristics.Our simulation and theoretical results fully demonstrate this interesting phenomenon.Our new interpretation mechanism provides not only the potential mathematical calculation principle of neural network but also an effective way to accurately match and predict human brain or animal activities,which can further expand and enrich the interpretable mechanism of artificial neural network in the future.

Special issue:Reliability management of complex system

Nowadays,reliability is moving toward interdisciplinary research with ever-increasing connotations for full life-cycle system management,including system design,analysis,modeling,test,operation,optimization,etc.Meanwhile,complex systems,such as transportation system,power system,communication system and other various critical infrastructure systems,have posed a big challenge,which attracts great attention both in theory and application.Characterized by nonlin ear interaction,emerge nt response,and high dime nsional coupling,the complex systems are in the face of extremely high uncertainty and vulnerability.