Optimization by a Team of Independent Automata with Adaptation

This paper continues our work [1-3] on the search and optimization capabilities of teams of independent automata. We shall present numerical data on some mechanisms of adaptation of optimizing automata. These mechanisms represent attempts to overcome the basic contradiction, inherent in any search system, between the desire to maximize the speed of search for the proper action (i.e., minimize the number of purposeless actions) and the desire to maintain a sufficiently large repertoire of such actions. Such attempts, induced by biological associations, involve the introduction of inhibitions and excitations that are produced by the system itself in response to its estimate of the effect of these actions (this estimate can be termed emotional), and also the introduction of the estimates (which accumulate during the operation of the system) of the individual automata taking part in the process.     

Quantitative Inference by Qualitative Semantic Knowledge Mining with Bayesian Model Averaging

In this paper, we consider the problem of performing quantitative Bayesian inference and model averaging based on a set of qualitative statements about relationships. Statements are transformed into parameter constraints which are imposed onto a set of Bayesian networks. Recurrent relationship structures are resolved by unfolding in time to Dynamic Bayesian networks. The approach enables probabilistic inference by model averaging, i.e. it allows to predict probabilistic quantities from a set of qualitative constraints without probability assignment on the model parameters. Model averaging is performed by Monte Carlo integration techniques. The method is applied to a problem in a molecular medical context: We show how the rate of breast cancer metastasis formation can be predicted based solely on a set of qualitative biological statements about the involvement of proteins in metastatic processes.     

Adaptation of Bayesian Models for Single-Channel Source Separation and its Application to Voice/Music Separation in Popular Songs

Probabilistic approaches can offer satisfactory solutions to source separation with a single channel, provided that the models of the sources match accurately the statistical properties of the mixed signals. However, it is not always possible to train such models. To overcome this problem, we propose to resort to an adaptation scheme for adjusting the source models with respect to the actual properties of the signals observed in the mix. In this paper, we introduce a general formalism for source model adaptation which is expressed in the framework of Bayesian models. Particular cases of the proposed approach are then investigated experimentally on the problem of separating voice from music in popular songs. The obtained results show that an adaptation scheme can improve consistently and significantly the separation performance in comparison with nonadapted models.     

基于隐马尔可夫模型边缘表示的多尺度边缘检测

基于增强图像特征空间的特性分析,提出一种新的图像边缘表示模型——隐马尔可夫模型,该模型把边缘作为一个可观测序列,认为它是人们根据感官知识(不可观测的状态量)所观察到的具有一定空间关系的像素集合;通过小波变换,得到多尺度增强图像;特包含在低分辨率图像边缘中的全局信息进行综合,作为高分辨率图像中隐马尔可夫边缘模型的训练数据,使用Viterbi算法完成高分辨率图像中边缘点的搜索,从而,使得多分辨率边缘检测对先验知识的依赖大大降低,同时降低了计算复杂度。边缘的隐马尔可夫模型表示也为多分辨率数据融合提供了一种有效的工具,实验表明该模型是合理的、可行的。     

XBSG－6／50型自动跟踪补偿的消弧线圈

ＸＢＳＧ－６／５０型自动跟踪补偿的消弧线圈，由单片机控制，不仅能在电网正常运行时自动跟踪电网的电容电流、而且在发生单相接地故障后，也能自动跟踪电网的电容电流．使单相接地电流最小。     

独立分量重建模型的手写数字字符识别

针对在手写字符识别中,因于书写习惯和风格的不同,造成字符模式不稳定的问题,利用独立分量对信号重建的良好性能,采用独立分量分析的方法抽取字符独立特征,并建立字符重建模型;通过对重建模型的误差分析进行字符识别;对美国国家邮政局USPS(US Postal Service database)字库中全部数字字符完整的识别实验,证实了算法的鲁棒性和准确性．     

A Bayesian/Information Theoretic Model of Learning to Learn via Multiple Task Sampling

A Bayesian model of learning to learn by sampling from multiple tasks is presented. The multiple tasks are themselves generated by sampling from a distribution over an environment of related tasks. Such an environment is shown to be naturally modelled within a Bayesian context by the concept of an objective prior distribution. It is argued that for many common machine learning problems, although in general we do not know the true (objective) prior for the problem, we do have some idea of a set of possible priors to which the true prior belongs. It is shown that under these circumstances a learner can use Bayesian inference to learn the true prior by learning sufficiently many tasks from the environment. In addition, bounds are given on the amount of information required to learn a task when it is simultaneously learnt with several other tasks. The bounds show that if the learner has little knowledge of the true prior, but the dimensionality of the true prior is small, then sampling multiple tasks is highly advantageous. The theory is applied to the problem of learning a common feature set or equivalently a low-dimensional-representation (LDR) for an environment of related tasks.     

Data Mining by Means of Binary Representation: A Model for Similarity and Clustering

In this paper we outline a new method for clustering that is based on a binary representation of data records. The binary database relates each entity to all possible attribute values (domain) that entity may assume. The resulting binary matrix allows for similarity and clustering calculation by using the positive (1 bits) of the entity vector. We formulate two indexes: Pair Similarity Index (PSI) to measure similarity between two entities and Group Similarity Index (GSI) to measure similarity within a group of entities. A threshold factor for each attribute domain is defined that is dependent on the domain but independent of the number of entities in the group. The similarity measure provides simplicity of storage and efficiency of calculation. A comparison of our similarity index to other indexes is made. Experiments with sample data indicate a 48% improvement of group similarity over standard methods pointing to the potential and merit of the binary approach to clustering and data mining.     

Fault detection in dynamic systems by a Fuzzy/Bayesian network formulation

In this paper the fault detection problem is solved using an alternative methodology based on a fuzzy/Bayesian strategy combining a Bayesian network and the fuzzy set theory. The new important issue in this proposed methodology is to address uncertainties in the input of the Bayesian Network. This contribution is possible since the fuzzy set theory is used as the knowledge representation. To illustrate the technique, the fault detection problem in induction machine stator-winding is considered. Specifically, the faults in the induction machine stator-winding are detected by a state change of stator current. Simulation results are presented to illustrate the advance of the proposed methodology when compared to standard Bayesian network.     

Model of Representation and Acquisition of New Knowledge by an Autonomous Intelligent Robot Based on the Logic of Conditionally Dependent Predicates

Journal of Computer and Systems Sciences International - The model for the representation of declarative and procedural knowledge of an autonomous intelligent robot is developed without reference...     

用类/微驱动模型开发DSP视频驱动程序

集成外设的增多,软件应用复杂性进一步的增加,导致开发外设驱动程序所需的工作量大大增加.TI公司提出了类/微驱动模型的驱动程序结构.实践结果表明,采用类/微驱动模型进行驱动程序设计后,应用软件可以复用绝大部分相似设备的驱动程序,从而极大地提高了驱动程序的开发效率.     

利用ActiveX技术实现B/S模式的电厂实时数据综合查询

介绍了使用Delphi的ActiveX技术实现B／S模式的电厂实时数据综合查询系统的相关技术,并给出具体示例。系统采用Delphi5.0,IIS4.0及FrontPage2000来完成。