The importance of service level management has come to the fore in recent years as computing power becomes more and more of a commodity. In order to present a consistently high quality of service systems must be rigorously analysed, even before implementation, and monitored to ensure these goals can be achieved. The tools and algorithms found in performability analysis offer a potentially ideal method to formally specify and analyse performance and reliability models. This thesis examines Markov reward models, a formalism based on continuous time Markov chains, and it's usage in the generation and analysis of service levels. The particular solution technique we employ in this thesis is model checking, using Continuous Reward Logic as a means to specify requirement and constraints on the model. We survey the current tools available allowing model checking to be performed on Markov reward models. Specifically we extended the Erlangen-Twente Markov Chain Checker to be able to solve Markov reward models by taking advantage of the Duality theorem of Continuous Stochastic Reward Logic, of which Continuous Reward Logic is a sub-logic. We are also concerned with the specification techniques available for Markov reward models, which have in the past merely been extensions to the available specification techniques for continuous time Markov chains. We implement a production rule system using Ruby, a high level language, and show the advantages gained by using it's native interpreter and language features in order to cut down on implementation time and code size. The limitations inherent in Markov reward models are discussed and we focus on the issue of zero reward states. Previous algorithms used to remove zero reward states, while preserving the numerical properties of the model, could potentially alter it's logical properties. We propose algorithms based on analysing the continuous reward logic requirement beforehand to determine whether a zero reward state can be removed safely as well as an approach based on substitution of zero reward states. We also investigate limitations on multiple reward structures and the ability to solve for both time and reward. Finally we perform a case study on a Beowulf parallel computing cluster using Markov reward models and the ETMCC tool, demonstrating their usefulness in the implementation of performability analysis and the determination of the service levels that can be offered by the cluster to it's users.

The study of human evolution centres, to a large extent, around the study of fossil morphology, including the comparison and interpretation of these remains within the context of what is known about morphological variation within living species. However, many fossils suffer from environmentally caused damage (taphonomic distortion) which hinders any such interpretation: fossil material may be broken and fragmented while the weight and motion of overlaying sediments can cause their plastic distortion. To date, a number of studies have focused on the reconstruction of such taphonomically damaged specimens. These studies have used myriad approaches to reconstruction, including thin plate spline methods, mirroring, and regression-based approaches. The efficacy of these techniques remains to be demonstrated, and it is not clear how different parameters (e.g., sample sizes, landmark density, etc.) might effect their accuracy. In order to partly address this issue, this thesis examines three techniques used in the virtual reconstruction of fossil remains by statistical or geometrical means: mean substitution, thin plate spline warping (TPS), and multiple linear regression. These methods are compared by reconstructing the same sample of individuals using each technique. Samples drawn from Homo sapiens, Pan troglodytes, Gorilla gorilla, and various hominin fossils are reconstructed by iteratively removing then estimating the landmarks. The testing determines the methods' behaviour in relation to the extant of landmark loss (i.e., amount of damage), reference sample sizes (this being the data used to guide the reconstructions), and the species of the population from which the reference samples are drawn (which may be different to the species of the damaged fossil). Given a large enough reference sample, the regression-based method is shown to produce the most accurate reconstructions. Various parameters effect this: when using small reference samples drawn from a population of the same species as the damaged specimen, thin plate splines is the better method, but only as long as there is little damage. As the damage becomes severe (missing 30% of the landmarks, or more), mean substitution should be used instead: thin plate splines are shown to have a rapid error growth in relation to the amount of damage. When the species of the damaged specimen is unknown, or it is the only known individual of its species, the smallest reconstruction errors are obtained with a regression-based approach using a large reference sample drawn from a living species. Testing shows that reference sample size (combined with the use of multiple linear regression) is more important than morphological similarity between the reference individuals and the damaged specimen. The main contribution of this work are recommendations to the researcher on which of the three methods to use, based on the amount of damage, number of reference individuals, and species of the reference individuals.

This dissertation builds upon the ideas introduced by Liben-Nowell and Kleinberg in The Link Prediction Problem for Social Networks [42]. Link prediction is the problem of predicting between which unconnected nodes in a graph a link will form next, based on the current structure of the graph. The following research contributions are made: • Highlighting the difference between the link prediction and link detection problems, which have been implicitly regarded as identical in current research. Despite hidden links and forming links having very highly significant differing metric values, they could not be distinguished from each other by a machine learning system using traditional metrics in an initial experiment. However, they could be distinguished from each other in a "simple" network (one where traditional metrics can be used for prediction successfully) using a combination of new graph analysis approaches. • Defining temporal metric statistics by combining traditional statistical measures with measures commonly employed in financial analysis and traditional social network analysis. These metrics are calculated over time for a sequence of sociograms. It is shown that some of the temporal extensions of traditional metrics increase the accuracy of link prediction. • Defining traditional metrics using different radii to those at which they are normally calculated. It is shown that this approach can increase the individual prediction accuracy of certain metrics, marginally increase the accuracy of a group of metrics, and greatly increase metric computation speed without sacrificing information content by computing metrics using smaller radii. It also solves the “distance-three task” (that common neighbour metrics cannot predict links between nodes at a distance greater than three). • Showing that the combination of local and temporal approaches to link prediction can lead to very high prediction accuracies. Furthermore in “complex” networks (ones where traditional metrics cannot be used for prediction successfully) local and temporal metrics become even more useful.

This (cumulative) dissertation is concerned with mechanisms and models of information processing and transmission by individual neurons and small neural assemblies. In this document, I first provide historical context for these ideas and highlight similarities and differences to related concepts from machine learning and neuromorphic engineering. With this background, I then discuss the four main themes of my work, namely dendritic filtering and delays, homeostatic plasticity and adaptation, rate-coding with spiking neurons, and spike-timing based alternatives to rate-coding. The content of this discussion is in large part derived from several of my own publications included in Appendix C, but it has been extended and revised to provide a more accessible and broad explanation of the main ideas, as well as to show their inherent connections. I conclude that fundamental differences remain between our understanding of information processing and transmission in machine learning on the one hand and theoretical neuroscience on the other, which should provide a strong incentive for further interdisciplinary work on the domain boundaries between neuroscience, machine learning and neuromorphic engineering.

The monitoring of continuous phenomena like temperature, air pollution, precipitation, soil moisture etc. is of growing importance. Decreasing costs for sensors and associated infrastructure increase the availability of observational data. These data can only rarely be used directly for analysis, but need to be interpolated to cover a region in space and/or time without gaps. So the objective of monitoring in a broader sense is to provide data about the observed phenomenon in such an enhanced form. Notwithstanding the improvements in information and communication technology, monitoring always has to function under limited resources, namely: number of sensors, number of observations, computational capacity, time, data bandwidth, and storage space. To best exploit those limited resources, a monitoring system needs to strive for efficiency concerning sampling, hardware, algorithms, parameters, and storage formats. In that regard, this work proposes and evaluates solutions for several problems associated with the monitoring of continuous phenomena. Synthetic random fields can serve as reference models on which monitoring can be simulated and exactly evaluated. For this purpose, a generator is introduced that can create such fields with arbitrary dynamism and resolution. For efficient sampling, an estimator for the minimum density of observations is derived from the extension and dynamism of the observed field. In order to adapt the interpolation to the given observations, a generic algorithm for the fitting of kriging parameters is set out. A sequential model merging algorithm based on the kriging variance is introduced to mitigate big workloads and also to support subsequent and seamless updates of real-time models by new observations. For efficient storage utilization, a compression method is suggested. It is designed for the specific structure of field observations and supports progressive decompression. The unlimited diversity of possible configurations of the features above calls for an integrated approach for systematic variation and evaluation. A generic tool for organizing and manipulating configurational elements in arbitrary complex hierarchical structures is proposed. Beside the root mean square error (RMSE) as crucial quality indicator, also the computational workload is quantified in a manner that allows an analytical estimation of execution time for different parallel environments. In summary, a powerful framework for the monitoring of continuous phenomena is outlined. With its tools for systematic variation and evaluation it supports continuous efficiency improvement.