No description available.

Designing new microprocessors is a time consuming task. Architects rely on slow simulators to evaluate performance and a significant proportion of the design space has to be explored before an implementation is chosen. This process becomes more time consuming when compiler optimisations are also considered. Once the architecture is selected, a new compiler must be developed and tuned. What is needed are techniques that can speedup this whole process and develop a new optimising compiler automatically. This thesis proposes the use of machine-learning techniques to address architecture/compiler co-design. First, two performance models are developed and are used to efficiently search the design space of amicroarchitecture. These models accurately predict performance metrics such as cycles or energy, or a tradeoff of the two. The first model uses just 32 simulations to model the entire design space of new applications, an order of magnitude fewer than state-of-the-art techniques. The second model addresses offline training costs and predicts the average behaviour of a complete benchmark suite. Compared to state-of-the-art, it needs five times fewer training simulations when applied to the SPEC CPU 2000 and MiBench benchmark suites. Next, the impact of compiler optimisations on the design process is considered. This has the potential to change the shape of the design space and improve performance significantly. A new model is proposed that predicts the performance obtainable by an optimising compiler for any design point, without having to build the compiler. Compared to the state-of-the-art, this model achieves a significantly lower error rate. Finally, a new machine-learning optimising compiler is presented that predicts the best compiler optimisation setting for any new program on any new microarchitecture. It achieves an average speedup of 1.14x over the default best gcc optimisation level. This represents 61% of the maximum speedup available, using just one profile run of the application.

The main objectives of this master thesis are to explore the effectiveness of nonlinear mixed effects model for longitudinal data. Mixed effect models allow to investigate the nature of relationship between the time-varying covariates and the response while also capturing the variations of subjects. I investigate the robustness of the longitudinal models by building up the complexity of the models starting from multiple linear models and ending up with additive nonlinear mixed models. I use a dataset where firms’ leverage are explained by four explanatory variables in addition to a grouping factor that is the firm factor. The models are compared using comparison statistics such as AIC, BIC and by a visual inspection of residuals. Likelihood ratio test has been used in some nested models only. The models are estimated by maximum likelihood and restricted maximum likelihood estimation. The most efficient model is the nonlinear mixed effects model which has lowest AIC and BIC. The multiple linear regression model failed to explain the relation and produced unrealistic statistics

In today’s world, ontologies are being widely used for data integration tasks and solving information heterogeneity problems on the web because of their capability in providing explicit meaning to the information. The growing need to resolve the heterogeneities between different information systems within a domain of interest has led to the rapid development of individual ontologies by different organizations. These ontologies designed for a particular task could be a unique representation of their project needs. Thus, integrating distributed and heterogeneous ontologies by finding semantic correspondences between their concepts has become the key point to achieve interoperability among different representations. In this thesis, an advanced instance-based ontology matching algorithm has been proposed to enable data integration tasks in ocean sensor networks, whose data are highly heterogeneous in syntax, structure, and semantics. This provides a solution to the ontology mapping problem in such systems based on machine-learning methods and string-based methods.

Face recognition research using automatic or semi-automatic techniques has emerged over the last two decades. One reason for growing interest in this topic is the wide range of possible applications for face recognition systems. Another reason is the emergence of affordable hardware, supporting digital photography and video, which have made the acquisition of high-quality and high resolution 2D images much more ubiquitous. However, 2D recognition systems are sensitive to subject pose and illumination variations and 3D face recognition which is not directly affected by such environmental changes, could be used alone, or in combination with 2D recognition. Recently with the development of more affordable 3D acquisition systems and the availability of 3D face databases, 3D face recognition has been attracting interest to tackle the limitations in performance of most existing 2D systems. In this research, we introduce a robust automated 3D Face recognition system that implements 3D data of faces with different facial expressions, hair, shoulders, clothing, etc., extracts features for discrimination and uses machine learning techniques to make the final decision. A novel system for automatic processing for 3D facial data has been implemented using multi stage architecture; in a pre-processing and registration stage the data was standardized, spikes were removed, holes were filled and the face area was extracted. Then the nose region, which is relatively more rigid than other facial regions in an anatomical sense, was automatically located and analysed by computing the precise location of the symmetry plane. Then useful facial features and a set of effective 3D curves were extracted. Finally, the recognition and matching stage was implemented by using cascade correlation neural networks and support vector machine for classification, and the nearest neighbour algorithms for matching. It is worth noting that the FRGC data set is the most challenging data set available supporting research on 3D face recognition and machine learning techniques are widely recognised as appropriate and efficient classification methods.

The space of compile-time transformations and or run-time options which can improve the performance of a given code is usually so large as to be virtually impossible to search in any practical time-frame. Thus, heuristics are leveraged which can suggest good but not necessarily best configurations. Unfortunately, since such heuristics are tightly coupled to processor architecture performance is not portable; heuristics must be tuned, traditionally manually, for each device in turn. This is extremely laborious and the result is often outdated heuristics and less effective optimisation. Ideally, to keep up with changes in hardware and run-time environments a fast and automated method to generate heuristics is needed. Recent works have shown that machine learning can be used to produce mathematical models or rules in their place, which is automated but not necessarily fast. This thesis proposes the use of active machine learning, sequential analysis, and active feature acquisition to accelerate the training process in an automatic way, thereby tackling this timely and substantive issue. First, a demonstration of the efficiency of active learning over the previously standard supervised machine learning technique is presented in the form of an ensemble algorithm. This algorithm learns a model capable of predicting the best processing device in a heterogeneous system to use per workload size, per kernel. Active machine learning is a methodology which is sensitive to the cost of training; specifically, it is able to reduce the time taken to construct a model by predicting how much is expected to be learnt from each new training instance and then only choosing to learn from those most profitable examples. The exemplar heuristic is constructed on average 4x faster than a baseline approach, whilst maintaining comparable quality. Next, a combination of active learning and sequential analysis is presented which reduces both the number of samples per training example as well as the number of training examples overall. This allows for the creation of models based on noisy information, sacrificing accuracy per training instance for speed, without having a significant affect on the quality of the final product. In particular, the runtime of high-performance compute kernels is predicted from code transformations one may want to apply using a heuristic which was generated up to 26x faster than with active learning alone. Finally, preliminary work demonstrates that an automated system can be created which optimises both the number of training examples as well as which features to select during training to further substantially accelerate learning, in cases where each feature value that is revealed comes at some cost.

No description available.

Face recognition research using automatic or semi-automatic techniques has emerged over the last two decades. One reason for growing interest in this topic is the wide range of possible applications for face recognition systems. Another reason is the emergence of affordable hardware, supporting digital photography and video, which have made the acquisition of high-quality and high resolution 2D images much more ubiquitous. However, 2D recognition systems are sensitive to subject pose and illumination variations and 3D face recognition which is not directly affected by such environmental changes, could be used alone, or in combination with 2D recognition. Recently with the development of more affordable 3D acquisition systems and the availability of 3D face databases, 3D face recognition has been attracting interest to tackle the limitations in performance of most existing 2D systems. In this research, we introduce a robust automated 3D Face recognition system that implements 3D data of faces with different facial expressions, hair, shoulders, clothing, etc., extracts features for discrimination and uses machine learning techniques to make the final decision. A novel system for automatic processing for 3D facial data has been implemented using multi stage architecture; in a pre-processing and registration stage the data was standardized, spikes were removed, holes were filled and the face area was extracted. Then the nose region, which is relatively more rigid than other facial regions in an anatomical sense, was automatically located and analysed by computing the precise location of the symmetry plane. Then useful facial features and a set of effective 3D curves were extracted. Finally, the recognition and matching stage was implemented by using cascade correlation neural networks and support vector machine for classification, and the nearest neighbour algorithms for matching. It is worth noting that the FRGC data set is the most challenging data set available supporting research on 3D face recognition and machine learning techniques are widely recognised as appropriate and efficient classification methods.

Current pavement performance prediction models are based on the parameters such as climate, traffic, environment, material properties, etc. while all these factors are playing important roles in the performance of pavements, the quality of construction and production are also as important as the other factors. The designed properties of Hot Mix Asphalt (HMA) pavements, known as flexible pavements, are subjected to change during production and construction stages. Therefore, most of the times the final product is not the exact reflection of the design. In almost any highway project, these changes are common and likely to occur from different sources, by various causes, and at any stage. These changes often have considerable impacts on the long-term performance of a project. The uncertainty of the traffic and environmental factors, as well as the variability of material properties and pavement structural systems, are obstacles for precise prediction of pavement performance. Therefore, it is essential to adopt a hybrid approach in pavement performance prediction and design; in which deterministic values work along with stochastic ones. Despite the advancement of technology, it is natural to observe variability during the production and construction stages of flexible pavements. Quality control programs are trying to minimize and control these variations and keep them at the desired levels. Utilizing the information gathered at the production and construction stages is beneficial for managers and researchers. This information enables performing analysis and investigations of pavements based on the as-produced and as-constructed values, rather than focusing on design values. This study describes a geo-relational framework to connect the pavement life-cycle information. This framework allows more intelligent and data-driven decisions for the pavements. The constructed geo-relational database can pave the way for artificial intelligence tools to help both researchers and practitioners having more accurate pavement design, quality control programs, and maintenance activities. This study utilizes data collected as part of quality control programs to develop more accurate deterioration and performance models. This data is not only providing the true perspective of actual measurements from different pavement properties but also answers how they are distributed over the length of the pavement. This study develops and utilizes different distribution functions of pavement properties and incorporate them into the general performance prediction models. These prediction models consist of different elements that are working together to produce an accurate and detailed prediction of performance. The model predicts occurrence and intensity of four common flexible pavement distresses; such as rutting, alligator, longitudinal and transverse cracking along with the total deterioration rate at different ages and locations of pavement based on material properties, traffic, and climate of a given highway. The uniqueness of the suggested models compared to the conventional pavement models in the literature is that; it carries out a multiscale and multiphysics approach which is believed to be essential for analyzing a complex system such as flexible pavements. This approach encompasses the discretization of the system into subsystems to employ the proper computational tools required to treat them. This approach is suitable for problems with a wide range of spatial and temporal scales as well as a wide variety of different coupled physical phenomena such as pavements. Moreover, the suggested framework in this study relies on using stochastic and machine learning techniques in the analysis along with the conventional deterministic methods. In addition, this study utilizes mechanical testing to provide better insights into the behavior of the pavement. A series of performance tests are conducted on field core samples with a variety of different material properties at different ages. These tests allow connecting the lab test results with the field performance survey and the material, environmental and loading properties. Moreover, the mix volumetrics extracted from the cores assisted verifying the distribution function models. Finally, the deterioration of flexible pavements as a result of four different distresses is individually investigated and based on the findings; different models are suggested. Dividing the roadway into small sections allowed predicting finer resolution of performance. These models are proposed to assist the highway agencies s in their pavement management process and quality control programs. The resulting models showed a strong ability to predict field performance at any age during the pavements service life. The results of this study highlighted the benefits of highway agencies in adopting a geo-relational framework for their pavement network. This study provides information and guidance to evolve towards data-driven pavement life cycle management consisted of quality pre-construction, quality during construction, and deterioration post-construction. / Civil Engineering

The rapid growth of virtualized data centers and cloud hosting services is making the management of physical resources such as CPU, memory, and I/O bandwidth in data center servers increasingly important. Server management now involves dealing with multiple dissimilar applications with varying Service-Level-Agreements (SLAs) and multiple resource dimensions. The multiplicity and diversity of resources and applications are rendering administrative tasks more complex and challenging. This thesis aimed to develop a framework and techniques that would help substantially reduce data center management complexity. We specifically addressed two crucial data center operations. First, we precisely estimated capacity requirements of client virtual machines (VMs) while renting server space in cloud environment. Second, we proposed a systematic process to efficiently allocate physical resources to hosted VMs in a data center. To realize these dual objectives, accurately capturing the effects of resource allocations on application performance is vital. The benefits of accurate application performance modeling are multifold. Cloud users can size their VMs appropriately and pay only for the resources that they need; service providers can also offer a new charging model based on the VMs performance instead of their configured sizes. As a result, clients will pay exactly for the performance they are actually experiencing; on the other hand, administrators will be able to maximize their total revenue by utilizing application performance models and SLAs. This thesis made the following contributions. First, we identified resource control parameters crucial for distributing physical resources and characterizing contention for virtualized applications in a shared hosting environment. Second, we explored several modeling techniques and confirmed the suitability of two machine learning tools, Artificial Neural Network and Support Vector Machine, to accurately model the performance of virtualized applications. Moreover, we suggested and evaluated modeling optimizations necessary to improve prediction accuracy when using these modeling tools. Third, we presented an approach to optimal VM sizing by employing the performance models we created. Finally, we proposed a revenue-driven resource allocation algorithm which maximizes the SLA-generated revenue for a data center.