A multiple intelligent agent system for credit risk prediction via an optimization of localized generalization error with diversity

Company bankruptcies cost billions of dollars in losses to banks each year. Thus credit risk prediction is a critical part of a bank＇s loan approval decision process. Traditional financial models for credit risk prediction are no longer adequate for describing today＇s complex relationship between the financial health and potential bankruptcy of a company. In this work, a multiple classifier system （embedded in a multiple intelligent agent system） is proposed to predict the financial health of a company. In our model, each individual agent （classifier） makes a prediction on the likelihood of credit risk based on only partial information of the company. Each of the agents is an expert, but has limited knowledge （represented by features） about the company. The decisions of all agents are combined together to form a final credit risk prediction. Experiments show that our model out-performs other existing methods using the benchmarking Compustat American Corporations dataset.     

A FRAMEWORK FOR REAL-VALUED CIPHER SYSTEMS

Most cipher systems designed thus far are binary-valued or integer-valued cipher systems. Their security relies on the assumption that one-way functions exist. Though the existence of one-way functions has not been proved yet, most cryptographic researchers believe that one-way functions exist. In addition, many candidates for one-way functions have been proposed. Therefore, the key step for developing real-valued cipher systems is to define real one-way functions and to propose candidates for them. In this paper, based on computational complexity theory over the real field, we give two definitions of real one-way functions; one is for digital one-way functions and the other is for general one-way functions. Candidates for these two classes of one-way functions are also proposed. Moreover, we present two examples to demonstrate that the candidates for both digital one-way functions and general one-way functions can be applied to construct secure real-valued cipher systems.     

NECESSARY AND SUFFICIENT CONDITION FOR GLOBAL CONTROLLABILITY OF A CLASS OF AFFINE NONLINEAR SYSTEMS

In this paper, we investigate the global controllability of a class of n-dimensional affine nonlinear systems with n- 1 controls and constant control matrix. A necessary and sufficient condition for its global controllability has been obtained by using the methods recently developed. Furthermore, we generalize the above result to a class of affine nonlinear systems with a block-triangular-like structure. Finally, we will give three examples to show the applications of our results.     

基于状态反馈的广义双线性系统H∞容错控制

针对非线性系统的容错控制问题,以广义双线性系统为研究对象,基于非线性系统H∞控制的概念,提出了利用状态反馈控制器实现广义双线性系统容错控制的方法,给出了状态反馈控制器的存在条件和设计过程。设计的状态反馈控制器在无执行器故障和存在执行器故障的情况下,都能保证闭环系统渐近稳定,且闭环输入输出信号满足H∞性能指标,从而实现系统的容错控制。最后的仿真结果证明了该方法的有效性。     

EXISTENCE AND UNIQUENESS AND STABILITY OF SOLUTIONS FOR STOCHASTIC IMPULSIVE SYSTEMS

This paper studies the existence, uniqueness, and stability of solutions for stochastic impulsive systems. By employing Lyapunov-like functions, some sufficient conditions of the global existence, uniqueness, and stability of solutions for stochastic impulsive systems are established. Furthermore, the results are specialized to the case of linear stochastic impulsive systems. Finally, some examples are given to illustrate the applications of our theory.     

INFORMATION CHARACTERIZATION OF COMMUNICATION CHANNELS FOR SYSTEM IDENTIFICATION

This paper studies identification of systems in which the system output is quantized, transmitted through a digital communication channel, and observed afterwards. The concept of the CR Ratio is introduced to characterize impact of communication channels on identification. The relationship between the CR Ratio and Shannon channel capacity is discussed. Identification algorithms are further developed when the channel error probability is unknown.     

复杂性视野中的生态城市系统建设

对复杂性理论的兴起、生态城市及其系统复杂性进行了分析总结,运用复杂性理论,提出了生态城市系统建设相应的建设对策。     

论系统论在工程伦理研究中的运用

现代工程活动是复杂的社会活动，其中涉及到许多复杂的伦理问题，工程伦理问题实际上已成为我们时代的一个亟待研究的新课题。采用系统论的基本原理，对工程伦理问题的整体性、层次性、相关性、开放性等系统特性进行分析，提出如何运用系统论方法促进工程伦理问题研究，希望从整体和最优化的角度，来探索和研究工程伦理问题。     

基于Agent的卫星航天系统建模与仿真方法研究

使用基于Agent的建模与仿真（ABMS）研究卫星航天系统,以解决对其总体功能和应用效能进行评估的问题。首先,介绍了卫星航天系统的概念和组成,并对其复杂性作了分析。然后,从应用角度介绍了Agent模型所应具备的特性,并指出了ABMS在传统建模与仿真方法的基础上还需开展的额外工作。最后,以一个火灾卫星系统为例进行了实现,证明了此方法应用的可行性和有效性。