En la actualidad se están realizando diversos esfuerzos para realizar la visión fu-turista de las redes de telecomunicación autogestionadas. La gestión basada en políticas ha sido reconocida como una herramienta potencial para habilitar esta visión. Mayorita-riamente, ésta técnica ha sido reconocida como proveedora de flexibilidad, adaptabilidad y soporte para asignar recursos, controlar Calidad de Servicio y seguridad, de una manera automática y de acuerdo a reglas administrativas. Adicionalmente, se ha considerado que la gestión basada en políticas proveería tal flexibilidad en tiempo de ejecución y como resultado de cambios en la red, interacciones entre usuarios, aplicaciones y disponibilidad de recursos. A pesar de enormes esfuerzos realizados con lenguajes de especificación de políticas, arquitecturas de gestión en diversos dominios y estandarización, la gestión ba-sada en políticas aún no es una realidad. Una de las razones para la reticencia en su utili-zación es la dificultad para analizar políticas que garanticen estabilidad en el sistema. Además de la problemática asociada a la gestión de conflictos entre políticas, otro obstá-culo para su utilización es la dificultad de derivar políticas ejecutables alineadas a objeti-vos administrativos o a otras políticas de alto nivel. Este último es el problema del refi namiento de políticas.Esta Tesis aborda el problema crítico de refinamiento de políticas. Damos una vi-sión completa del proceso de refinamiento, desde el análisis formal hasta su realización práctica, identificando los elementos que intervienen en cada paso de tal proceso. Inicialmente, proponemos un marco de trabajo para refinamiento de políticas ba-sado en Lógica Lineal Temporal, una lógica estándar que permite el análisis en sistemas reactivos. Esta técnica es utilizada para representar políticas a diferentes niveles jerár-quicos de abstracción. Acto seguido desarrollamos mecanismos que habilitan la obten-ción de políticas ejecutables a partir de ciertos requerimientos mediante la utilización de técnicas de exploración de estados basados en Lógica Lineal Temporal. Adicionalmente, aclaramos e identificamos las actividades y tareas de gestión de las partes administrativas durante el ciclo de vida de un sistema de gestión basado en políticas, desde la perspectiva del proceso de refinamiento de políticas. Esta Tesis presenta también directrices para abordar el proceso de refinamiento de políticas en contextos de gestión de red. Damos un paso adelante en la materialización de este proceso mediante la utilización de propiedades estructurales inherentes a sistemas de gestión de red. Proveemos, en fin, una metodología para aplicar los conceptos introduci-dos en el marco de trabajo desarrollado en esta Tesis en sistemas de gestión de red. En esta Tesis también realizamos un proceso de refinamiento de políticas com-pleto. Detallamos la realización de tal proceso en una solución exitosa de gestión basada en políticas. Tomando como base el dominio de Gestión de Calidad de Servicio, aclara-mos y presentamos las implicaciones del problema de refinamiento en este dominio de aplicación. / Current research efforts are being directed to commit with the long-term view of self management properties for telecommunications networks. One of the key approaches that have been recognised as an enabler of such a view is policy-based management. Pol-icy-based management has been mostly acknowledged as a methodology that provides flexibility, adaptability and support to automatically assign network resources, control Quality of Service and security, by considering administratively specified rules. The hype of policy-based management was to commit with these features in run-time as a result of changeable network conditions resulting from the interactions of users, applications and existing resources. Despite enormous efforts with policy languages, management archi-tectures using policy in different application domains, standardisation and industrial ef-forts, policy-based management is still not a reality. One reason behind the reticence for its use is the difficulty to analyse policies that guarantee configuration stability. In addi-tion to policy conflict analysis, a key issue for this reticence is the need to derive en-forceable policies from high level administrative goals or from higher level policies, namely the policy refinement process.This Thesis deals with the critical nature of addressing the policy refinement problem. We provide a holistic view of this process, from formal analysis to its practical realisation, identifying the key elements involved in each step of such critical process. We initially propose a policy refinement framework relying on Linear Temporal Logic (LTL), a standard logic that allows analysis of reactive systems. Based on the for-mer logic, we lay down the process of representing policies at different levels of abstrac-tion. Following on with this, we develop the mechanisms that enable the abstraction of enforceable policies from hierarchical requirements in a fully automatic manner, making use of Linear Temporal Logic based state exploration techniques. In addition, we clarify and identify the activities and management tasks that the administrative parties should carry out during the life cycle of the policy-based management system, from the perspec tive of the policy refinement process.This Thesis provides the guidelines to address policy refinement in network management contexts. Concretely, we take one step ahead in the materialisation of the policy refinement process by exploiting inherent containment properties of network management systems. For this purpose we provide the methodology to apply the concepts introduced in the policy refinement framework developed in this Thesis in the above context.In this Thesis we also execute a complete and rather detailed policy refinement process for a successful policy-based management solution. Taking the intra-domain Quality of Service Management application domain as background, we clarify and pre-sent the implications of the policy refinement problem in such a concrete application do-main.

Influenza A and B viruses are major causes of seasonal flu epidemics each year. Hemagglutinin (HA) mediates the binding of virus to host cell and the fusion with host membrane. The crystal of HA in complex with antibody that reveals the mechanism by which antibody recognizes HA may not diffract to high resolution, thereby preventing the accurate interpretation of the structural model. The application of normal mode refinement that aims for improving the structure quality at the low resolution is tested. These studies provide some guidelines for future refinement of HA-antibody complex structures. By comparing the residues constituting the base of the receptor binding site of influenza A and B virus HAs, it is found that they share some similarities, except for a Phe at position 95 of influenza B virus hemagglutinin (BHA) versus Tyr in of influenza A virus hemagglutinin (AHA). The recombinant protein BHA containing the F95Y mutation exhibits the increased receptor binding affinity and specificity. However, recombinant viruses with the Phe95Tyr mutation show lower erythrocyte agglutination titer and decreased binding abilities with different cell lines. The replication of the Phe95Tyr mutant virus in mice is also attenuated. These data suggest that the increased receptor binding ability of HA alone is not advantageous to the pathogenesis of the viruses. The structure of BHA2 (a portion of BHA near the C-terminus) at the post-fusion state has been determined to 2.45 Å resolution. This protein forms a hairpin-like conformation rich in -helices. About 70 residues from the N-terminus is a three-stranded coiled coil, and the remaining of the protein packs in anti-parallel against the groove formed by the central helices. In the post-fusion state of BHA2, the helix converted from the B-loop in pre-fusion state contacts the C-terminal fragment of this protein with more hydrophobic interactions as compared to AHA2. This structure illustrates the distinct stabilization strategy employed by BHA2 to form a post-fusion state that resembles that for AHA2. These studies will further the understanding of BHA with respect to its role in receptor binding ability and fusion.

When solving partial differential equations using finite difference methods on structured meshes, adaptive refinement can be used to increase the accuracy of the solution in an efficient manner. When implementing solvers using structured adaptive mesh refinement for modern parallel computer systems, an important task is the partitioning of the grid hierarchy over the available processors. The Meta-partitioner is an autonomic framework which can dynamically select between a large number of grid-partitioning algorithms at run time. In this thesis we investigate which modifications that are necessary in order to connect the Meta-partitioner to the existing SAMR-framework Chombo, and begin the process of performing this connection. We conclude that although significant changes to both Chombo and the Meta-partitioner are necessary, a connection definitely seams feasible. We estimate that that the major work of the connection has been done, and that with the experience gained from this project, the continuation is straightforward. We also connect a patch-based partitioning algorithm to Chombo and evaluate it for the first time as part of a real SAMR-based simulation. The results are promising and we conclude that it is a viable candidate for inclusion in the Meta-partitioner.

none

Immune molecules, including complement proteins, C1q and C3, have emerged as critical mediators of synaptic refinement and plasticity. Complement proteins localize to synapses and refine the developing retinogeniculate system via C3-dependent microglial phagocytosis of synapses. Retinal ganglion cells (RGCs) express C1q, the initiating protein of the classical complement cascade, during retinogeniculate refinement; however, the signals controlling C1q expression and function remain elusive. RGCs grown in the presence of astrocytes significantly upregulated C1q compared to controls, implicating an astrocyte-derived factor in neuronal C1q expression. A major goal of my dissertation research was to identify the signals that regulate C1q expression and function in the developing visual system. In this study, I have identified transforming growth factor beta \((TGF-\beta)\), an astrocyte-secreted cytokine, as both necessary and sufficient for C1q expression in RGCs through an activity-dependent mechanism. Specific disruption of retinal \(TGF-\beta\) signaling resulted in a significant reduction in the deposition of C1q and downstream C3 at retinogeniculate synapses and significant synaptic refinement defects in the retinogeniculate system. Microglia engulfment of RGC inputs in the lateral geniculate nucleus (LGN) was also significantly reduced in retinal \(TGF\beta\)RII KOs, phenocopying the engulfment defects observed in C1q KOs, C3 KOs, and CR3 KOs. Interestingly, in C1q KOs and retinal \(TGF\beta\)RII KOs, microglia also failed to preferentially engulf less active inputs when retinal activity was manipulated, suggesting that retinal activity and \(TGF-\beta\) signaling cooperatively regulate complement mediated synaptic refinement. In support of this hypothesis, blocking spontaneous activity in RGC cultures significantly reduced C1q upregulation by \(TGF-\beta\). Moreover, manipulating spontaneous retinal activity in vivo modulated C1q expression levels in RGCs and C1q deposition in the LGN. Together these findings support a model in which retinal activity and \(TGF-\beta\) signaling control expression and local release of C1q in the LGN to regulate microglia-mediated, complement-dependent synaptic pruning. These results provide mechanistic insight into synaptic refinement and, potentially, pathological synapse loss which occurs in the early stages of neurodegenerative diseases concurrently with aberrant complement expression and reactive gliosis.

Magistro darbe įvardijamas Hipotekos registro kaip organizacijos valdymo tobulinimas darbuotojų ir respondentų nuomonių aspektu. Darbe pateikiama įvairių Lietuvos ir užsienio autorių nuomonių analizė. Analizuojama organizacijos samprata, valdymo kriterijai, vadovavimo stiliai, mokymo reikšmė, mokymosi ir kvalifikacijos tobulinimo motyvai, darbo sąlygos, darbo užmokesčio bei skatinimo veiksniai. Darbe palyginamos darbuotojų ir klientų nuomonės dėl aptarnavimo lygio, atliekama darbuotojų nuomonės analizė organizacijos valdymo tobulinimo veiksniams nustatyti. Darbo rezultatai bus geri tada, kai darbuotojams ne tik bus išaiškinta, ko iš jų norima, bet ir išklausoma jų, suteikiama tai, ko jie nori. Pasitvirtina teiginys, kad darbuotojai gerai, uoliai ir kokybiškai dirba gaudami už tai atitinkamą darbo atlygį, nuolat skatinami ir suteikiant jiems galimybę kelti kvalifikaciją ir tobulėti. / This Master`s paper presents management refinement of a Mortgage Registry from employees’ and respondents’ point of view. It contains a deep analysis of Lithuanian and other foreign authors’ opinions. The analysis is oriented around the conception of an organization, management criteria, management style, and meaning of training, learning and qualification refinement motives, work conditions, salary and its stimulation. This Master`s paper compares employee and customer opinions regarding the level of service provided as well as provides an analysis of employee opinion to determine what factors influence management development within an organization. The results will be satisfying only when employees are not only aware of what employer wants from them, but also when someone listens to employees and gives them what employees need. The affirmation that employees do a quality job when they get paid appropriate salary, when they are always encouraged and given an opportunity to develop and grow, truly holds.

The jarosite group of minerals is part of the alunite supergroup, which consists of more than 40 different mineral species that have the general formula AB3(TO4)2(OH, H2O)6. There is extensive solid-solution in the A, B and T sites within the alunite supergroup. Jarosite group minerals are common in acid mine waste and there is evidence of jarosite existing on Mars. Members of the jarosite - natrojarosite – hydronium jarosite (K,Na, H3O)Fe3(SO4)2(OH)6 solid-solution series were synthesized and investigated by Rietveld analysis of X-ray powder diffraction data. The synthesized samples have full iron occupancy, where in many previous studies there was significant vacancies in the B site. Well-defined trends can be seen in the unit cell parameters, bond lengths A – O and Fe - O across the solid-solution series in the synthetic samples. Based on unit cell parameters many natural samples appear to have full iron occupancy and correlate well with the synthetic samples from this study. In addition, the infrared spectra of the samples were analyzed. The atomic structure of ammoniojarosite, (NH4)Fe3(SO4)2(OH)6, has been solved using single-crystal X-ray diffraction to wR 3.64% and R 1.4%. The atomic coordinates of the hydrogen atoms of the NH4 group have been located and it was found that the ammonium group has two different orientations with equal probability. Samples in the ammoniojarosite – hydronium jarosite solid-solution series were synthesized and analyzed using powder X-ray diffraction and Rietveld refinement. It was found that an incomplete solid-solution series exists between jarosite and plumbojarosite, Pb[Fe3(SO4)2(OH)6]2, based on experimental and mineralogical data. At the studied synthesis conditions, lead solubility in jarosite is extremely limited with occupancy of 2% in the potassium site. Increased Pb in the iv starting solution resulted in no increased substitution of Pb into jarosite, but an increased substitution of H3O+. The stable isotope (H) geochemistry of hydronium jarosite, (H3O,K)Fe3(SO4)2(O,OH)6, and the effect that the presence of hydronium in the crystal structure has on exchange rates of stable isotope values of jarosite with hydronium substitution has been investigated in this study. / Thesis (Ph.D, Geological Sciences & Geological Engineering) -- Queen's University, 2008-05-07 18:21:45.136

No description available.

The logical framework LF and its metalogic Twelf can be used to encode and reason about a wide variety of logics, languages, and other deductive systems in a formal, machine-checkable way. Recent studies have shown that ML-like languages can profitably be extended with a notion of subtyping called refinement types. A refinement type discipline uses an extra layer of term classification above the usual type system to more accurately capture certain properties of terms. I propose that adding refinement types to LF is both useful and practical. To support the claim, I exhibit an extension of LF with refinement types called LFR,work out important details of itsmetatheory, delineate a practical algorithmfor refinement type reconstruction, andpresent several case studies that highlight the utility of refinement types for formalized mathematics. In the end I find that refinement types and LF are a match made in heaven: refinements enable many rich new modes of expression, and the simplicity of LF ensures that they come at a modest cost.

Artificial neural networks (ANNs) are essentially a 'black box' technology. The lack of an explanation component prevents the full and complete exploitation of this form of machine learning. During the mid 1990's the field of 'rule extraction' emerged. Rule extraction techniques attempt to derive a human comprehensible explanation structure from a trained ANN. Andrews et.al. (1995) proposed the following reasons for extending the ANN paradigm to include a rule extraction facility: * provision of a user explanation capability * extension of the ANN paradigm to 'safety critical' problem domains * software verification and debugging of ANN components in software systems * improving the generalization of ANN solutions * data exploration and induction of scientific theories * knowledge acquisition for symbolic AI systems An allied area of research is that of 'rule refinement'. In rule refinement an initial rule base, (i.e. what may be termed `prior knowledge') is inserted into an ANN by prestructuring some or all of the network architecture, weights, activation functions, learning rates, etc. The rule refinement process then proceeds in the same way as normal rule extraction viz (1) train the network on the available data set(s); and (2) extract the `refined' rules. Very few ANN techniques have the capability to act as a true rule refinement system. Existing techniques, such as KBANN, (Towell & Shavlik, (1993), are limited in that the rule base used to initialize the network must be a nearly complete, and the refinement process is limited to modifying antecedents. The limitations of existing techniques severely limit their applicability to real world problem domains. Ideally, a rule refinement technique should be able to deal with incomplete initial rule bases, modify antecedents, remove inaccurate rules, and add new knowledge by generating new rules. The motivation for this research project was to develop such a rule refinement system and to investigate its efficacy when applied to both nearly complete and incomplete problem domains. The premise behind rule refinement is that the refined rules better represent the actual domain theory than the initial domain theory used to initialize the network. The hypotheses tested in this research include: * that the utilization of prior domain knowledge will speed up network training, * produce smaller trained networks, * produce more accurate trained networks, and * bias the learning phase towards a solution that 'makes sense' in the problem domain. In 1998 Geva, Malmstrom, & Sitte, (1998) described the Local Cluster (LC) Neural Net. Geva et.al. (1998) showed that the LC network was able to learn / approximate complex functions to a high degree of accuracy. The hidden layer of the LC network is comprised of basis functions, (the local cluster units), that are composed of sigmoid based 'ridge' functions. In the General form of the LC network the ridge functions can be oriented in any direction. We describe RULEX, a technique designed to provide an explanation component for its underlying Local Cluster ANN through the extraction of symbolic rules from the weights of the local cluster units of the trained ANN. RULEX exploits a feature, ie, axis parallel ridge functions, of the Restricted Local Cluster (Geva , Andrews & Geva 2002), that allow hyper-rectangular rules of the form IF ∀ 1 ≤ i ≤ n : xi ∈ [ xi lower , xi upper ] THEN pattern belongs to the target class to be easily extracted from local functions that comprise the hidden layer of the LC network. RULEX is tested on 14 applications available in the public domain. RULEX results are compared with a leading machine learning technique, See5, with RULEX generally performing as well as See5 and in some cases outperforming See5 in predictive accuracy. We describe RULEIN, a rule refinement technique that allows symbolic rules to be converted into the parameters that define local cluster functions. RULEIN allows existing domain knowledge to be captured in the architecture of a LC ANN thus facilitating the first phase of the rule refinement paradigm. RULEIN is tested on a variety of artificial and real world problems. Experimental results indicate that RULEIN is able to satisfy the first requirement of a rule refinement technique by correctly translating a set of symbolic rules into a LC ANN that has the same predictive bahaviour as the set of rules from which it was constructed. Experimental results also show that in the cases where a strong domain theory exists, initializing an LC network using RULEIN generally speeds up network training, produces smaller, more accurate trained networks, with the trained network properly representing the underlying domain theory. In cases where a weak domain theory exists the same results are not always apparent. Experiments with the RULEIN / LC / RULEX rule refinement method show that the method is able to remove inaccurate rules from the initial knowledge base, modify rules in the initial knowledge base that are only partially correct, and learn new rules not present in the initial knowledge base. The combination of RULEIN / LC / RULEX thus is shown to be an effective rule refinement technique for use with a Restricted Local Cluster network.