随机半范模上等度连续的典则模同态族的特征

给出了随机半范模上等度连续的典则模同态族的一个待征——随机半范模上强凸的等度连续的典则模同态族可以被一个性质较好的连续随机半范数所控制．     

A Population-Based Approach to the Resource-Constrained Project Scheduling Problem

We present a population-based approach to the RCPSP. The procedure has two phases. The first phase handles the initial construction of a population of schedules and these are then evolved until high quality solutions are obtained. The evolution of the population is driven by the alternative application of an efficient improving procedure for locally improving the use of resources, and a mechanism for combining schedules that blends scatter search and path relinking characteristics. The objective of the second phase is to explore in depth those vicinities near the high quality schedules. Computational experiments on the standard j120 set, generated using ProGen, show that our algorithm produces higher quality solutions than state-of-the-art heuristics for the RCPSP in an average time of less than five seconds.     

An Algorithm for Constructing Single Processor Static–Dynamic Schedules

An extended approach is proposed for using algorithms to find the maximum flow in a transport network when constructing schedules for performing work with interruptions in a new class of problems of creating static–dynamic schedules. This is accomplished by projecting realtime systems with the architecture of integrated module avionics. The results from an experimental study confirm the high accuracy of the proposed algorithm.     

In Memoriam: Michael (Miki) Neumann (1946–2011)

Kernel canonical correlation analysis (KCCA) is a procedure for assessing the relationship between two sets of random variables when the classical method, canonical correlation analysis (CCA), fails because of the nonlinearity of the data. The KCCA method is mostly used in machine learning, especially for information retrieval and text mining. Because the data is often represented with non-negative numbers, we propose to incorporate the non-negativity restriction directly into the KCCA method. Similar restrictions have been studied in relation to the classical CCA and called restricted canonical correlation analysis (RCCA), so that we call the proposed method restricted kernel canonical correlation analysis (RKCCA). We also provide some possible approaches for solving the optimization problem to which our method translates. The motivation for introducing RKCCA is given in Section 2.     

Adaptation in a stochastic multi-resources chemostat model

We are interested in modeling the Darwinian evolution resulting from the interplay of phenotypic variation and natural selection through ecological interactions, in the specific scales of the biological framework of adaptive dynamics. Adaptive dynamics so far has been put on a rigorous footing only for direct competition models (Lotka–Volterra models) involving a competition kernel which describes the competition pressure from one individual to another one. We extend this to a multi-resources chemostat model, where the competition between individuals results from the sharing of several resources which have their own dynamics. Starting from a stochastic birth and death process model, we prove that, when advantageous mutations are rare, the population behaves on the mutational time scale as a jump process moving between equilibrium states (the polymorphic evolution sequence of the adaptive dynamics literature). An essential technical ingredient is the study of the long time behavior of a chemostat multi-resources dynamical system. In the small mutational steps limit this process in turn gives rise to a differential equation in phenotype space called canonical equation of adaptive dynamics. From this canonical equation and still assuming small mutation steps, we prove a rigorous characterization of the evolutionary branching points.     

Learning from Hints

We present a systematic method for incorporating prior knowledge (hints) into the learning-from-examples paradigm. The hints are represented in a canonical form that is compatible with descent techniques for learning. All the hints are fed to the learning process in the form of examples, and examples of the function are treated on equal footing with the rest of the hints. During learning, examples from different hints are selected for processing according to a fixed or adaptive schedule. Fixed schedules specify the relative emphasis of each hint, and adaptive schedules are based on how well each hint has been learned so far. We discuss adaptive minimization which is based on estimates of the overall learning error.     

Bridging the gap between self schedules and feasible schedules in staff scheduling

Every company that has employees working on irregular schedules must deal with the difficult and time consuming problem of creating feasible schedules for the employees. We introduce an algorithm that takes a partial schedule created by requests from employees and creates feasible schedule where most of the employee’s requests are unchanged, while still making sure that rules and regulations are not violated. The algorithm is based on independent modules, which can be executed in any order, and each module tries to emulate some action taken by a staff manager. Our goal is to create a transparent and fair system that creates feasible schedules of high quality, but also a system where the employees can get an explanation and justification for every change that the algorithm makes to the employee requests. By emulating the actions of staff managers, the algorithm is easily understood by staff managers and, using detailed logs of any action, make any decision easy to explain to the employees. We will present the algorithm and show results from four real world companies and institutions. The results show that a simple module based heuristic can get good results and create fair and feasible schedules that encourage employees to participate in the self-scheduling process.     

基于均匀设计抽样遗传算法求解背包问题

众所周知,遗传算法的运行机理及特点是具有定向制导的随机搜索技术,其定向制导的原则是:导向以高适应度模式为祖先的"家族"方向.以此结论为基础,利用均匀设计抽样的理论和方法,对遗传算法中的交叉操作进行了重新设计,给出了一个新的GA算法,称之为均匀设计抽样遗传算法.最后将均匀设计抽样遗传算法应用于求解背包问题,并与简单遗传算...     

Upper bounds on complexity of Frege proofs with limited use of certain schemata

The paper considers a commonly used axiomatization of the classical propositional logic and studies how different axiom schemata in this system contribute to proof complexity of the logic. The existence of a polynomial bound on proof complexity of every statement provable in this logic is a well-known open question. The axiomatization consists of three schemata. We show that any statement provable using unrestricted number of axioms from the first of the three schemata and polynomially-bounded in size set of axioms from the other schemata, has a polynomially-bounded proof complexity. In addition, it is also established, that any statement, provable using unrestricted number of axioms from the remaining two schemata and polynomially-bounded in size set of axioms from the first scheme, also has a polynomially-bounded proof complexity.     

Emergence of order from chaos by cooperative non-biased learning between random bit-strings

Numerical experiments show that non-biased learning between families of independent and random bit-strings causes order. A parallel distributed learning between these bit-strings is performed by a neural network of the type pattern associator. The system allows emergence of some order in the learning matrix when a non-linear process is used, while a linear learning is unable to break the learning-matrix randomness. This neural network is in fact a complex and non-linear dynamical system, and consequently is able to self-organize order from chaos. It is also a model of collective proto-cognition that would describe biological evolution in species by cooperative learning. Our model suggests that the cause of evolution towards order in complex systems, can be just the learning process.     

Extended canonical algebras and Fuchsian singularities

The authors introduce a new class of finite dimensional algebras called extended canonical, and determine the shape of their derived categories. Extended canonical algebras arise from a canonical algebra ?? by onepoint extension or coextension by an indecomposable projective module. Our main results concern the case of negative Euler characteristic of the corresponding weighted projective line ${\mathbb{X}}$ ; more specifically we establish, for a base field of arbitrary characteristic, a link to the Fuchsian singularity R of ${\mathbb{X}}$ which for the base field of complex numbers is isomorphic to an algebra of automorphic forms. By means of a recent result of Orlov we show that the triangulated category of the graded singularities of R (in the sense of Buchweitz and Orlov) admits a tilting object whose endomorphism ring is the corresponding extended canonical algebra. Of particular interest are those cases where the attached Coxeter transformation has spectral radius one. A K-theoretic analysis then shows that this happens exactly for 38 cases including Arnold??s 14 exceptional unimodal singularities. The paper is related to recent independent work by Kajiura, Saito and Takahashi.     

关于随机赋范空间与随机内积空间的某些基本理论(英文)

首先提出随机度量空间定义的另一个提法,这提法不仅等价于原始的定义而且也使随机度量空间自动归入广义度量空间的框架,也考虑了关于拓扑结构的某些新的问题;循着同样的思路,对随机赋范空间的定义也作了新的处理并同时简化了随机赋范模的定义．其次本文也证明了一个Ｅ－范空间的商空间等距同构于一个典型的Ｅ－范空间;进一步,在概率赋范空间的框架下证明了一个概率赋伪范空间是伪内积生成空间的充要条件是它等距同构于一个Ｅ－内积空间,这回答了Ｃ．Ａｌｓｉｎａ与Ｂ．Ｓｃｈｗｅｉｚｅｒ等人新近提出的公开问题．最后,本文转向了它的中心部分──关于随机内积空间的研究,对随机内积空间中的特有且复杂的正交性作较系统的讨论,论证了只有几乎处处正交性才是唯一合理的正交性概念,在此基础上本文尤其将Ｇ．Ｓｔａｍｐａｃｃｈｉａ的在众多学科中都有多种用途的一般投影定理（或称变分不等式解存在性定理）以适当形式推广到完备实随机内积模上．     

Periodic Modules and Quantum Groups

We prove that the elements A_\leqslant defined by Lusztig in a completion of the periodic module actually live in the periodic module (in the type A case). In order to prove this, we compare, using the Schur duality, these elements with the Kashiwara canonical basis of an integrable module.     

The influence of random interactions and decision heuristics on norm evolution in social networks

In this paper we explore the effect that random social interactions have on the emergence and evolution of social norms in a simulated population of agents. In our model agents observe the behaviour of others and update their norms based on these observations. An agent’s norm is influenced by both their own fixed social network plus a second random network that is composed of a subset of the remaining population. Random interactions are based on a weighted selection algorithm that uses an individual’s path distance on the network to determine their chance of meeting a stranger. This means that friends-of-friends are more likely to randomly interact with one another than agents with a higher degree of separation. We then contrast the cases where agents make highest utility based rational decisions about which norm to adopt versus using a Markov Decision process that associates a weight with the best choice. Finally we examine the effect that these random interactions have on the evolution of a more complex social norm as it propagates throughout the population. We discover that increasing the frequency and weighting of random interactions results in higher levels of norm convergence and in a quicker time when agents have the choice between two competing alternatives. This can be attributed to more information passing through the population thereby allowing for quicker convergence. When the norm is allowed to evolve we observe both global consensus formation and group splintering depending on the cognitive agent model used.     

Random duality

The purpose of this paper is to provide a random duality theory for the further development of the theory of random conjugate spaces for random normed modules. First, the complicated stratification structure of a module over the algebra L(μ, K) frequently makes our investigations into random duality theory considerably different from the corresponding ones into classical duality theory, thus in this paper we have to first begin in overcoming several substantial obstacles to the study of stratification structure on random locally convex modules. Then, we give the representation theorem of weakly continuous canonical module homomorphisms, the theorem of existence of random Mackey structure, and the random bipolar theorem with respect to a regular random duality pair together with some important random compatible invariants.     

Modeling scale-free neurodynamics using neuropercolation approach

Critical properties of dynamical models of neural populations are studied. Based on the classical work of Renyi-Erdos on the evolution of random graphs, a new class of random cellular automata models called neuropercolation has been introduced. We show the emergence of phase transitions in neuropercolation models at critical combination of several control parameters, including the level of external gain and noise, the density of long-range axonal connections (small-world phenomenon), and the sparseness of feedback between excitatory and inhibitory neural populations. Noise level and structural properties of the cortical tissue has been used to control the critical exponent, starting from white noise (slope 0) far away from criticality. The results show that scale-free power spectral density characterizes the dynamics near criticality, where exponent with a power exponent approaching –2. The results are interpreted in the context of recent experimental findings on the dynamics and structure of the cortex. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Gaussian free fields for mathematicians

The d-dimensional Gaussian free field (GFF), also called the (Euclidean bosonic) massless free field, is a d-dimensional-time analog of Brownian motion. Just as Brownian motion is the limit of the simple random walk (when time and space are appropriately scaled), the GFF is the limit of many incrementally varying random functions on d-dimensional grids. We present an overview of the GFF and some of the properties that are useful in light of recent connections between the GFF and the Schramm–Loewner evolution. Partially supported by NSF grant DMS0403182.     

A vision architecture for unconstrained and incremental learning of multiple categories

We present an integrated vision architecture capable of incrementally learning several visual categories based on natural hand-held objects. Additionally we focus on interactive learning, which requires real-time image processing methods and a fast learning algorithm. The overall system is composed of a figure-ground segregation part, several feature extraction methods and a life-long learning approach combining incremental learning with category-specific feature selection. In contrast to most visual categorization approaches, where typically each view is assigned to a single category, we allow labeling with an arbitrary number of shape and color categories. We also impose no restrictions on the viewing angle of presented objects, relaxing the common constraint on canonical views.     

Image compression with a dynamic autoassociative neural network

Image compression using neural networks has been attempted with some promise. Among the architectures, feedforward backpropagation networks (FFBPN) have been used in several attempts. Although it is demonstrated that using the mean quadratic error function is equivalent to applying the Karhunen-Loeve transformation method, promise still arises from directed learning possibilities, generalization abilities and performance of the network once trained. In this paper we propose an architecture and an improved training method to attempt to solve some of the shortcomings of traditional data compression systems based on feedforward neural networks trained with backpropagation—the dynamic autoassociation neural network (DANN).The successful application of neural networks to any task requires proper training of the network. In this research, this issue is taken as the main consideration in the design of DANN. We emphasize the convergence of the learning process proposed by DANN. This process provides an escape mechanism, by adding neurons in a random state, to avoid the local minima trapping seen in traditional PFBPN. Also, DANN's training algorithm constrains the error for every pattern to an allowed interval to balance the training for every pattern, thus improving the performance rates in recognition and generalization. The addition of these two mechanisms to DANN's training algorithm has the result of improving the final quality of the images processed by DANN.The results of several tasks presented to DANN-based compression are compared and contrasted with the performance of an FFBPN-based system applied to the same task. These results indicate that DANN is superior to FFBPN when applied to image compression.     

Stochastic Integrals and Evolution Equations with Gaussian Random Fields

The paper studies stochastic integration with respect to Gaussian processes and fields. It is more convenient to work with a field than a process: by definition, a field is a collection of stochastic integrals for a class of deterministic integrands. The problem is then to extend the definition to random integrands. An orthogonal decomposition of the chaos space of the random field, combined with the Wick product, leads to the Itô-Skorokhod integral, and provides an efficient tool to study the integral, both analytically and numerically. For a Gaussian process, a natural definition of the integral follows from a canonical correspondence between random processes and a special class of random fields. Also considered are the corresponding linear stochastic evolution equations.