Adaptive learning of dynamic Bayesian networks with changing structures by detecting geometric structures of time series

A dynamic Bayesian network (DBN) is one of popular approaches for relational knowledge discovery such as modeling relations or dependencies, which change over time, between variables of a dynamic system. In this paper, we propose an adaptive learning method (autoDBN) to learn DBNs with changing structures from multivariate time series. In autoDBN, segmentation of time series is achieved first through detecting geometric structures transformed from time series, and then model regions are found from the segmentation by designed finding strategies; in each found model region, a DBN model is established by existing structure learning methods; finally, model revisiting is developed to refine model regions and improve DBN models. These techniques provide a special mechanism to find accurate model regions and discover a sequence of DBNs with changing structures, which are adaptive to changing relations between multivariate time series. Experimental results on simulated and real time series show that autoDBN is very effective in finding accurate/reasonable model regions and gives lower error rates, outperforming the switching linear dynamic system method and moving window method.

Automatic Construction and Verification of Isotopy Invariants

We extend our previous study of the automatic construction of isomorphic classification theorems for algebraic domains by considering the isotopy equivalence relation. Isotopism is an important generalisation of isomorphism, and is studied by mathematicians in domains such as loop theory. This extension was not straightforward, and we had to solve two major technical problems, namely, generating and verifying isotopy invariants. Concentrating on the domain of loop theory, we have developed three novel techniques for generating isotopic invariants, by using the notion of universal identities and by using constructions based on subblocks. In addition, given the complexity of the theorems that verify that a conjunction of the invariants form an isotopy class, we have developed ways of simplifying the problem of proving these theorems. Our techniques employ an interplay of computer algebra, model generation, theorem proving, and satisfiability-solving methods. To demonstrate the power of the approach, we generate isotopic classification theorems for loops of size 6 and 7, which extend the previously known enumeration results. This work was previously beyond the capabilities of automated reasoning techniques. The author’s work was supported by EPSRC MathFIT grant GR/S31099.     

A Mechanized Proof of the Basic Perturbation Lemma

We present a complete mechanized proof of the result in homological algebra known as basic perturbation lemma. The proof has been carried out in the proof assistant Isabelle, more concretely, in the implementation of higher-order logic (HOL) available in the system. We report on the difficulties found when dealing with abstract algebra in HOL, and also on the ongoing stages of our project to give a certified version of some of the algorithms present in the Kenzo symbolic computation system. J. Aransay was partially supported by Ministerio de Educación y Ciencia, MTM2006/06513, and by Gobierno de La Rioja ANGI2005/19 and J. Rubio was partially supported by Ministerio de Educación y Ciencia, MTM2006/06513, and by Gobierno de La Rioja ANGI2005/19.     

Approximate factorization of multivariate polynomials using singular value decomposition

We describe the design, implementation and experimental evaluation of new algorithms for computing the approximate factorization of multivariate polynomials with complex coefficients that contain numerical noise. Our algorithms are based on a generalization of the differential forms introduced by W. Ruppert and S. Gao to many variables, and use singular value decomposition or structured total least squares approximation and Gauss–Newton optimization to numerically compute the approximate multivariate factors. We demonstrate on a large set of benchmark polynomials that our algorithms efficiently yield approximate factorizations within the coefficient noise even when the relative error in the input is substantial (10⁻³).     

基于偏序简化的并发系统验证

构件交互风格和交互协议的描述与验证是基于构件的分布式系统开发的基础和关键,而构件交互协议是一种典型的分布式并发系统.传统的方法难以解决系统建模和验证中的所谓的状态爆炸问题.偏序简化是应用迹的概念,对模型进行化简并且对模型进行死锁验证.但这样的验证重点放在了Petri网模型上,而没有涉及进程代数模型,所验证的只是模型是否有死锁状态.而以通信系统演算CCS为代表的进程代数,因其概念简洁,可用的数学工具丰富,在分布式并发系统的规范、分析、设计和验证方面获得了广泛应用.对此,提出将偏序规约应用于进程代数模型,给出基于进程代数模型的偏序简化算法,并提出利用进程代数模型偏序简化算法来验证安全性的方法.     

Finite element formulation for dynamics of planar flexible multi-beam system

In some previous geometric nonlinear finite element formulations, due to the use of axial displacement, the contribution of all the elements lying between the reference node of zero axial displacement and the element to the foreshortening effect should be taken into account. In this paper, a finite element formulation is proposed based on geometric nonlinear elastic theory and finite element technique. The coupling deformation terms of an arbitrary point only relate to the nodal coordinates of the element at which the point is located. Based on Hamilton principle, dynamic equations of elastic beams undergoing large overall motions are derived. To investigate the effect of coupling deformation terms on system dynamic characters and reduce the dynamic equations, a complete dynamic model and three reduced models of hub-beam are prospected. When the Cartesian deformation coordinates are adopted, the results indicate that the terms related to the coupling deformation in the inertia forces of dynamic equations have small effect on system dynamic behavior and may be neglected, whereas the terms related to coupling deformation in the elastic forces are important for system dynamic behavior and should be considered in dynamic equation. Numerical examples of the rotating beam and flexible beam system are carried out to demonstrate the accuracy and validity of this dynamic model. Furthermore, it is shown that a small number of finite elements are needed to obtain a stable solution using the present coupling finite element formulation.     

Geometric nonlinear effects on the planar dynamics of a pivoted flexible beam encountering a point-surface impact

Flexible-body modeling with geometric nonlinearities remains a hot topic of research by applications in multibody system dynamics undergoing large overall motions. However, the geometric nonlinear effects on the impact dynamics of flexible multibody systems have attracted significantly less attention. In this paper, a point-surface impact problem between a rigid ball and a pivoted flexible beam is investigated. The Hertzian contact law is used to describe the impact process, and the dynamic equations are formulated in the floating frame of reference using the assumed mode method. The two important geometric nonlinear effects of the flexible beam are taken into account, i.e., the longitudinal foreshortening effect due to the transverse deformation, and the stress stiffness effect due to the axial force. The simulation results show that good consistency can be obtained with the nonlinear finite element program ABAQUS/Explicit if proper geometric nonlinearities are included in the floating frame formulation. Specifically, only the foreshortening effect should be considered in a pure transverse impact for efficiency, while the stress stiffness effect should be further considered in an oblique case with much more computational effort. It also implies that the geometric nonlinear effects should be considered properly in the impact dynamic analysis of more general flexible multibody systems.     

Inductive semimodules and the vector modules over them

The notion of an inductive semimodule over an ordered *-semiring is introduced and some related properties are investigated. Inductive semimodules are extensions of several important algebraic structures such as Kleene modules, Kleene algebras and inductive *-semirings. We prove that an inductive semimodule over an ordered *-semiring K is a Kleene module if and only if K is a Kleene algebra. Moreover, we establish that the vector module of an inductive semimodule over an ordered Conway semiring is again an inductive semimodule over the matrix semiring. Consequently, in an inductive semimodule over an ordered Conway semiring, least solutions to linear inequation systems can be denoted by linear expressions, avoiding the least fixed point operator. In addition, we also introduce a related notion called weak inductive semimodules, and propose several open problems on them.

Cognitive-Mathematical Approaches for Evaluating Architectural Contextual Fit

The main goal of this study is to apply a scientific quantitative approach to the investigation of contextual fit. This is approached mathematically within the framework of cognitive science and research on categorization and prototypes. Two experiments investigated two leading mathematical-cognitive approaches for explaining people’s judgment of contextual fit of a new building with an architectural/urban context: prototype approach and feature frequency approach. The basic concept is that people represent the built environment via architectural prototypes and/or frequencies of encountered architectural features. In the first experiment, a group of twelve participants performed rank order tasks on artificially created architectural patterns, for the purpose of psychological scaling. Perceptual distances among all patterns were mathematically determined. In the second experiment, three groups of architectural patterns were constructed to represent assumed architectural contexts. The prototype of each context was mathematically determined according to prototype cognitive model, and based on the distances calculated in the first experiment. Fifty-six students participated in the main experiment, in which they rank ordered a group of fifteen architectural patterns in terms of contextual fit to each of the three architectural contexts. Participants’ rank order data of the fifteen patterns were regressed on both the perceptual distances from prototypes, and numbers of features shared with each architectural context. Results indicated that both prototype and feature frequency approaches significantly accounted for important portions of participants’ judgments. However, participants tended to prefer one approach to the other according to context composition. Results have implications for both research on utilizing cognitive-mathematical models in architectural research and on urban design guidelines and control.