Refine
Year of publication
Document Type
- Doctoral Thesis (129) (remove)
Language
- English (129) (remove)
Keywords
- Maschinelles Lernen (7)
- Machine Learning (6)
- Antwortmengenprogrammierung (5)
- Answer Set Programming (3)
- answer set programming (3)
- machine learning (3)
- Algorithmen (2)
- Algorithms (2)
- Computersicherheit (2)
- Coq (2)
Institute
- Institut für Informatik und Computational Science (129) (remove)
A central insight from psychological studies on human eye movements is that eye movement patterns are highly individually characteristic. They can, therefore, be used as a biometric feature, that is, subjects can be identified based on their eye movements. This thesis introduces new machine learning methods to identify subjects based on their eye movements while viewing arbitrary content. The thesis focuses on probabilistic modeling of the problem, which has yielded the best results in the most recent literature. The thesis studies the problem in three phases by proposing a purely probabilistic, probabilistic deep learning, and probabilistic deep metric learning approach. In the first phase, the thesis studies models that rely on psychological concepts about eye movements. Recent literature illustrates that individual-specific distributions of gaze patterns can be used to accurately identify individuals. In these studies, models were based on a simple parametric family of distributions. Such simple parametric models can be robustly estimated from sparse data, but have limited flexibility to capture the differences between individuals. Therefore, this thesis proposes a semiparametric model of gaze patterns that is flexible yet robust for individual identification. These patterns can be understood as domain knowledge derived from psychological literature. Fixations and saccades are examples of simple gaze patterns. The proposed semiparametric densities are drawn under a Gaussian process prior centered at a simple parametric distribution. Thus, the model will stay close to the parametric class of densities if little data is available, but it can also deviate from this class if enough data is available, increasing the flexibility of the model. The proposed method is evaluated on a large-scale dataset, showing significant improvements over the state-of-the-art. Later, the thesis replaces the model based on gaze patterns derived from psychological concepts with a deep neural network that can learn more informative and complex patterns from raw eye movement data. As previous work has shown that the distribution of these patterns across a sequence is informative, a novel statistical aggregation layer called the quantile layer is introduced. It explicitly fits the distribution of deep patterns learned directly from the raw eye movement data. The proposed deep learning approach is end-to-end learnable, such that the deep model learns to extract informative, short local patterns while the quantile layer learns to approximate the distributions of these patterns. Quantile layers are a generic approach that can converge to standard pooling layers or have a more detailed description of the features being pooled, depending on the problem. The proposed model is evaluated in a large-scale study using the eye movements of subjects viewing arbitrary visual input. The model improves upon the standard pooling layers and other statistical aggregation layers proposed in the literature. It also improves upon the state-of-the-art eye movement biometrics by a wide margin. Finally, for the model to identify any subject — not just the set of subjects it is trained on — a metric learning approach is developed. Metric learning learns a distance function over instances. The metric learning model maps the instances into a metric space, where sequences of the same individual are close, and sequences of different individuals are further apart. This thesis introduces a deep metric learning approach with distributional embeddings. The approach represents sequences as a set of continuous distributions in a metric space; to achieve this, a new loss function based on Wasserstein distances is introduced. The proposed method is evaluated on multiple domains besides eye movement biometrics. This approach outperforms the state of the art in deep metric learning in several domains while also outperforming the state of the art in eye movement biometrics.
The objective and motivation behind this research is to provide applications with easy-to-use interfaces to communities of deaf and functionally illiterate users, which enables them to work without any human assistance. Although recent years have witnessed technological advancements, the availability of technology does not ensure accessibility to information and communication technologies (ICT). Extensive use of text from menus to document contents means that deaf or functionally illiterate can not access services implemented on most computer software. Consequently, most existing computer applications pose an accessibility barrier to those who are unable to read fluently. Online technologies intended for such groups should be developed in continuous partnership with primary users and include a thorough investigation into their limitations, requirements and usability barriers. In this research, I investigated existing tools in voice, web and other multimedia technologies to identify learning gaps and explored ways to enhance the information literacy for deaf and functionally illiterate users. I worked on the development of user-centered interfaces to increase the capabilities of deaf and low literacy users by enhancing lexical resources and by evaluating several multimedia interfaces for them. The interface of the platform-independent Italian Sign Language (LIS) Dictionary has been developed to enhance the lexical resources for deaf users. The Sign Language Dictionary accepts Italian lemmas as input and provides their representation in the Italian Sign Language as output. The Sign Language dictionary has 3082 signs as set of Avatar animations in which each sign is linked to a corresponding Italian lemma. I integrated the LIS lexical resources with MultiWordNet (MWN) database to form the first LIS MultiWordNet(LMWN). LMWN contains information about lexical relations between words, semantic relations between lexical concepts (synsets), correspondences between Italian and sign language lexical concepts and semantic fields (domains). The approach enhances the deaf users’ understanding of written Italian language and shows that a relatively small set of lexicon can cover a significant portion of MWN. Integration of LIS signs with MWN made it useful tool for computational linguistics and natural language processing. The rule-based translation process from written Italian text to LIS has been transformed into service-oriented system. The translation process is composed of various modules including parser, semantic interpreter, generator, and spatial allocation planner. This translation procedure has been implemented in the Java Application Building Center (jABC), which is a framework for extreme model driven design (XMDD). The XMDD approach focuses on bringing software development closer to conceptual design, so that the functionality of a software solution could be understood by someone who is unfamiliar with programming concepts. The transformation addresses the heterogeneity challenge and enhances the re-usability of the system. For enhancing the e-participation of functionally illiterate users, two detailed studies were conducted in the Republic of Rwanda. In the first study, the traditional (textual) interface was compared with the virtual character-based interactive interface. The study helped to identify usability barriers and users evaluated these interfaces according to three fundamental areas of usability, i.e. effectiveness, efficiency and satisfaction. In another study, we developed four different interfaces to analyze the usability and effects of online assistance (consistent help) for functionally illiterate users and compared different help modes including textual, vocal and virtual character on the performance of semi-literate users. In our newly designed interfaces the instructions were automatically translated in Swahili language. All the interfaces were evaluated on the basis of task accomplishment, time consumption, System Usability Scale (SUS) rating and number of times the help was acquired. The results show that the performance of semi-literate users improved significantly when using the online assistance. The dissertation thus introduces a new development approach in which virtual characters are used as additional support for barely literate or naturally challenged users. Such components enhanced the application utility by offering a variety of services like translating contents in local language, providing additional vocal information, and performing automatic translation from text to sign language. Obviously, there is no such thing as one design solution that fits for all in the underlying domain. Context sensitivity, literacy and mental abilities are key factors on which I concentrated and the results emphasize that computer interfaces must be based on a thoughtful definition of target groups, purposes and objectives.
Although it has become common practice to build applications based on the reuse of existing components or services, technical complexity and semantic challenges constitute barriers to ensuring a successful and wide reuse of components and services. In the geospatial application domain, the barriers are self-evident due to heterogeneous geographic data, a lack of interoperability and complex analysis processes.
Constructing workflows manually and discovering proper services and data that match user intents and preferences is difficult and time-consuming especially for users who are not trained in software development. Furthermore, considering the multi-objective nature of environmental modeling for the assessment of climate change impacts and the various types of geospatial data (e.g., formats, scales, and georeferencing systems) increases the complexity challenges.
Automatic service composition approaches that provide semantics-based assistance in the process of workflow design have proven to be a solution to overcome these challenges and have become a frequent demand especially by end users who are not IT experts. In this light, the major contributions of this thesis are:
(i) Simplification of service reuse and workflow design of applications for climate impact analysis by following the eXtreme Model-Driven Development (XMDD) paradigm.
(ii) Design of a semantic domain model for climate impact analysis applications that comprises specifically designed services, ontologies that provide domain-specific vocabulary for referring to types and services, and the input/output annotation of the services using the terms defined in the ontologies.
(iii) Application of a constraint-driven method for the automatic composition of workflows for analyzing the impacts of sea-level rise. The application scenario demonstrates the impact of domain modeling decisions on the results and the performance of the synthesis algorithm.
The main objective of this dissertation is to analyse prerequisites, expectations, apprehensions, and attitudes of students studying computer science, who are willing to gain a bachelor degree. The research will also investigate in the students’ learning style according to the Felder-Silverman model. These investigations fall in the attempt to make an impact on reducing the “dropout”/shrinkage rate among students, and to suggest a better learning environment.
The first investigation starts with a survey that has been made at the computer science department at the University of Baghdad to investigate the attitudes of computer science students in an environment dominated by women, showing the differences in attitudes between male and female students in different study years. Students are accepted to university studies via a centrally controlled admission procedure depending mainly on their final score at school. This leads to a high percentage of students studying subjects they do not want. Our analysis shows that 75% of the female students do not regret studying computer science although it was not their first choice. And according to statistics over previous years, women manage to succeed in their study and often graduate on top of their class. We finish with a comparison of attitudes between the freshman students of two different cultures and two different university enrolment procedures (University of Baghdad, in Iraq, and the University of Potsdam, in Germany) both with opposite gender majority.
The second step of investigation took place at the department of computer science at the University of Potsdam in Germany and analyzes the learning styles of students studying the three major fields of study offered by the department (computer science, business informatics, and computer science teaching). Investigating the differences in learning styles between the students of those study fields who usually take some joint courses is important to be aware of which changes are necessary to be adopted in the teaching methods to address those different students. It was a two stage study using two questionnaires; the main one is based on the Index of Learning Styles Questionnaire of B. A. Solomon and R. M. Felder, and the second questionnaire was an investigation on the students’ attitudes towards the findings of their personal first questionnaire. Our analysis shows differences in the preferences of learning style between male and female students of the different study fields, as well as differences between students with the different specialties (computer science, business informatics, and computer science teaching).
The third investigation looks closely into the difficulties, issues, apprehensions and expectations of freshman students studying computer science. The study took place at the computer science department at the University of Potsdam with a volunteer sample of students. The goal is to determine and discuss the difficulties and issues that they are facing in their study that may lead them to think in dropping-out, changing the study field, or changing the university. The research continued with the same sample of students (with business informatics students being the majority) through more than three semesters. Difficulties and issues during the study were documented, as well as students’ attitudes, apprehensions, and expectations. Some of the professors and lecturers opinions and solutions to some students’ problems were also documented. Many participants had apprehensions and difficulties, especially towards informatics subjects. Some business informatics participants began to think of changing the university, in particular when they reached their third semester, others thought about changing their field of study. Till the end of this research, most of the participants continued in their studies (the study they have started with or the new study they have changed to) without leaving the higher education system.
With the downscaling of CMOS technologies, the radiation-induced Single Event Transient (SET) effects in combinational logic have become a critical reliability issue for modern integrated circuits (ICs) intended for operation under harsh radiation conditions. The SET pulses generated in combinational logic may propagate through the circuit and eventually result in soft errors. It has thus become an imperative to address the SET effects in the early phases of the radiation-hard IC design. In general, the soft error mitigation solutions should accommodate both static and dynamic measures to ensure the optimal utilization of available resources. An efficient soft-error-aware design should address synergistically three main aspects: (i) characterization and modeling of soft errors, (ii) multi-level soft error mitigation, and (iii) online soft error monitoring. Although significant results have been achieved, the effectiveness of SET characterization methods, accuracy of predictive SET models, and efficiency of SET mitigation measures are still critical issues. Therefore, this work addresses the following topics: (i) Characterization and modeling of SET effects in standard combinational cells, (ii) Static mitigation of SET effects in standard combinational cells, and (iii) Online particle detection, as a support for dynamic soft error mitigation.
Since the standard digital libraries are widely used in the design of radiation-hard ICs, the characterization of SET effects in standard cells and the availability of accurate SET models for the Soft Error Rate (SER) evaluation are the main prerequisites for efficient radiation-hard design. This work introduces an approach for the SPICE-based standard cell characterization with the reduced number of simulations, improved SET models and optimized SET sensitivity database. It has been shown that the inherent similarities in the SET response of logic cells for different input levels can be utilized to reduce the number of required simulations. Based on characterization results, the fitting models for the SET sensitivity metrics (critical charge, generated SET pulse width and propagated SET pulse width) have been developed. The proposed models are based on the principle of superposition, and they express explicitly the dependence of the SET sensitivity of individual combinational cells on design, operating and irradiation parameters. In contrast to the state-of-the-art characterization methodologies which employ extensive look-up tables (LUTs) for storing the simulation results, this work proposes the use of LUTs for storing the fitting coefficients of the SET sensitivity models derived from the characterization results. In that way the amount of characterization data in the SET sensitivity database is reduced significantly.
The initial step in enhancing the robustness of combinational logic is the application of gate-level mitigation techniques. As a result, significant improvement of the overall SER can be achieved with minimum area, delay and power overheads. For the SET mitigation in standard cells, it is essential to employ the techniques that do not require modifying the cell structure. This work introduces the use of decoupling cells for improving the robustness of standard combinational cells. By insertion of two decoupling cells at the output of a target cell, the critical charge of the cell’s output node is increased and the attenuation of short SETs is enhanced. In comparison to the most common gate-level techniques (gate upsizing and gate duplication), the proposed approach provides better SET filtering. However, as there is no single gate-level mitigation technique with optimal performance, a combination of multiple techniques is required. This work introduces a comprehensive characterization of gate-level mitigation techniques aimed to quantify their impact on the SET robustness improvement, as well as introduced area, delay and power overhead per gate. By characterizing the gate-level mitigation techniques together with the standard cells, the required effort in subsequent SER analysis of a target design can be reduced. The characterization database of the hardened standard cells can be utilized as a guideline for selection of the most appropriate mitigation solution for a given design.
As a support for dynamic soft error mitigation techniques, it is important to enable the online detection of energetic particles causing the soft errors. This allows activating the power-greedy fault-tolerant configurations based on N-modular redundancy only at the high radiation levels. To enable such a functionality, it is necessary to monitor both the particle flux and the variation of particle LET, as these two parameters contribute significantly to the system SER. In this work, a particle detection approach based on custom-sized pulse stretching inverters is proposed. Employing the pulse stretching inverters connected in parallel enables to measure the particle flux in terms of the number of detected SETs, while the particle LET variations can be estimated from the distribution of SET pulse widths. This approach requires a purely digital processing logic, in contrast to the standard detectors which require complex mixed-signal processing. Besides the possibility of LET monitoring, additional advantages of the proposed particle detector are low detection latency and power consumption, and immunity to error accumulation.
The results achieved in this thesis can serve as a basis for establishment of an overall soft-error-aware database for a given digital library, and a comprehensive multi-level radiation-hard design flow that can be implemented with the standard IC design tools. The following step will be to evaluate the achieved results with the irradiation experiments.
TrainTrap
(2020)
Companies develop process models to explicitly describe their business operations. In the same time, business operations, business processes, must adhere to various types of compliance requirements. Regulations, e.g., Sarbanes Oxley Act of 2002, internal policies, best practices are just a few sources of compliance requirements. In some cases, non-adherence to compliance requirements makes the organization subject to legal punishment. In other cases, non-adherence to compliance leads to loss of competitive advantage and thus loss of market share. Unlike the classical domain-independent behavioral correctness of business processes, compliance requirements are domain-specific. Moreover, compliance requirements change over time. New requirements might appear due to change in laws and adoption of new policies. Compliance requirements are offered or enforced by different entities that have different objectives behind these requirements. Finally, compliance requirements might affect different aspects of business processes, e.g., control flow and data flow. As a result, it is infeasible to hard-code compliance checks in tools. Rather, a repeatable process of modeling compliance rules and checking them against business processes automatically is needed. This thesis provides a formal approach to support process design-time compliance checking. Using visual patterns, it is possible to model compliance requirements concerning control flow, data flow and conditional flow rules. Each pattern is mapped into a temporal logic formula. The thesis addresses the problem of consistency checking among various compliance requirements, as they might stem from divergent sources. Also, the thesis contributes to automatically check compliance requirements against process models using model checking. We show that extra domain knowledge, other than expressed in compliance rules, is needed to reach correct decisions. In case of violations, we are able to provide a useful feedback to the user. The feedback is in the form of parts of the process model whose execution causes the violation. In some cases, our approach is capable of providing automated remedy of the violation.
One of the main problems in machine learning is to train a predictive model from training data and to make predictions on test data. Most predictive models are constructed under the assumption that the training data is governed by the exact same distribution which the model will later be exposed to. In practice, control over the data collection process is often imperfect. A typical scenario is when labels are collected by questionnaires and one does not have access to the test population. For example, parts of the test population are underrepresented in the survey, out of reach, or do not return the questionnaire. In many applications training data from the test distribution are scarce because they are difficult to obtain or very expensive. Data from auxiliary sources drawn from similar distributions are often cheaply available. This thesis centers around learning under differing training and test distributions and covers several problem settings with different assumptions on the relationship between training and test distributions-including multi-task learning and learning under covariate shift and sample selection bias. Several new models are derived that directly characterize the divergence between training and test distributions, without the intermediate step of estimating training and test distributions separately. The integral part of these models are rescaling weights that match the rescaled or resampled training distribution to the test distribution. Integrated models are studied where only one optimization problem needs to be solved for learning under differing distributions. With a two-step approximation to the integrated models almost any supervised learning algorithm can be adopted to biased training data. In case studies on spam filtering, HIV therapy screening, targeted advertising, and other applications the performance of the new models is compared to state-of-the-art reference methods.
The programmable network envisioned in the 1990s within standardization and research for the Intelligent Network is currently coming into reality using IPbased Next Generation Networks (NGN) and applying Service-Oriented Architecture (SOA) principles for service creation, execution, and hosting. SOA is the foundation for both next-generation telecommunications and middleware architectures, which are rapidly converging on top of commodity transport services. Services such as triple/quadruple play, multimedia messaging, and presence are enabled by the emerging service-oriented IPMultimedia Subsystem (IMS), and allow telecommunications service providers to maintain, if not improve, their position in the marketplace. SOA becomes the de facto standard in next-generation middleware systems as the system model of choice to interconnect service consumers and providers within and between enterprises. We leverage previous research activities in overlay networking technologies along with recent advances in network abstraction, service exposure, and service creation to develop a paradigm for a service environment providing converged Internet and Telecommunications services that we call Service Broker. Such a Service Broker provides mechanisms to combine and mediate between different service paradigms from the two domains Internet/WWW and telecommunications. Furthermore, it enables the composition of services across these domains and is capable of defining and applying temporal constraints during creation and execution time. By adding network-awareness into the service fabric, such a Service Broker may also act as a next generation network-to-service element allowing the composition of crossdomain and cross-layer network and service resources. The contribution of this research is threefold: first, we analyze and classify principles and technologies from Information Technologies (IT) and telecommunications to identify and discuss issues allowing cross-domain composition in a converging service layer. Second, we discuss service composition methods allowing the creation of converged services on an abstract level; in particular, we present a formalized method for model-checking of such compositions. Finally, we propose a Service Broker architecture converging Internet and Telecom services. This environment enables cross-domain feature interaction in services through formalized obligation policies acting as constraints during service discovery, creation, and execution time.
In many applications one is faced with the problem of inferring some functional relation between input and output variables from given data. Consider, for instance, the task of email spam filtering where one seeks to find a model which automatically assigns new, previously unseen emails to class spam or non-spam. Building such a predictive model based on observed training inputs (e.g., emails) with corresponding outputs (e.g., spam labels) is a major goal of machine learning. Many learning methods assume that these training data are governed by the same distribution as the test data which the predictive model will be exposed to at application time. That assumption is violated when the test data are generated in response to the presence of a predictive model. This becomes apparent, for instance, in the above example of email spam filtering. Here, email service providers employ spam filters and spam senders engineer campaign templates such as to achieve a high rate of successful deliveries despite any filters. Most of the existing work casts such situations as learning robust models which are unsusceptible against small changes of the data generation process. The models are constructed under the worst-case assumption that these changes are performed such to produce the highest possible adverse effect on the performance of the predictive model. However, this approach is not capable to realistically model the true dependency between the model-building process and the process of generating future data. We therefore establish the concept of prediction games: We model the interaction between a learner, who builds the predictive model, and a data generator, who controls the process of data generation, as an one-shot game. The game-theoretic framework enables us to explicitly model the players' interests, their possible actions, their level of knowledge about each other, and the order at which they decide for an action. We model the players' interests as minimizing their own cost function which both depend on both players' actions. The learner's action is to choose the model parameters and the data generator's action is to perturbate the training data which reflects the modification of the data generation process with respect to the past data. We extensively study three instances of prediction games which differ regarding the order in which the players decide for their action. We first assume that both player choose their actions simultaneously, that is, without the knowledge of their opponent's decision. We identify conditions under which this Nash prediction game has a meaningful solution, that is, a unique Nash equilibrium, and derive algorithms that find the equilibrial prediction model. As a second case, we consider a data generator who is potentially fully informed about the move of the learner. This setting establishes a Stackelberg competition. We derive a relaxed optimization criterion to determine the solution of this game and show that this Stackelberg prediction game generalizes existing prediction models. Finally, we study the setting where the learner observes the data generator's action, that is, the (unlabeled) test data, before building the predictive model. As the test data and the training data may be governed by differing probability distributions, this scenario reduces to learning under covariate shift. We derive a new integrated as well as a two-stage method to account for this data set shift. In case studies on email spam filtering we empirically explore properties of all derived models as well as several existing baseline methods. We show that spam filters resulting from the Nash prediction game as well as the Stackelberg prediction game in the majority of cases outperform other existing baseline methods.
An increasing number of applications requires user interfaces that facilitate the handling of large geodata sets. Using virtual 3D city models, complex geospatial information can be communicated visually in an intuitive way. Therefore, real-time visualization of virtual 3D city models represents a key functionality for interactive exploration, presentation, analysis, and manipulation of geospatial data. This thesis concentrates on the development and implementation of concepts and techniques for real-time city model visualization. It discusses rendering algorithms as well as complementary modeling concepts and interaction techniques. Particularly, the work introduces a new real-time rendering technique to handle city models of high complexity concerning texture size and number of textures. Such models are difficult to handle by current technology, primarily due to two problems: - Limited texture memory: The amount of simultaneously usable texture data is limited by the memory of the graphics hardware. - Limited number of textures: Using several thousand different textures simultaneously causes significant performance problems due to texture switch operations during rendering. The multiresolution texture atlases approach, introduced in this thesis, overcomes both problems. During rendering, it permanently maintains a small set of textures that are sufficient for the current view and the screen resolution available. The efficiency of multiresolution texture atlases is evaluated in performance tests. To summarize, the results demonstrate that the following goals have been achieved: - Real-time rendering becomes possible for 3D scenes whose amount of texture data exceeds the main memory capacity. - Overhead due to texture switches is kept permanently low, so that the number of different textures has no significant effect on the rendering frame rate. Furthermore, this thesis introduces two new approaches for real-time city model visualization that use textures as core visualization elements: - An approach for visualization of thematic information. - An approach for illustrative visualization of 3D city models. Both techniques demonstrate that multiresolution texture atlases provide a basic functionality for the development of new applications and systems in the domain of city model visualization.
This thesis presents novel ideas and research findings for the Web of Data – a global data space spanning many so-called Linked Open Data sources. Linked Open Data adheres to a set of simple principles to allow easy access and reuse for data published on the Web. Linked Open Data is by now an established concept and many (mostly academic) publishers adopted the principles building a powerful web of structured knowledge available to everybody. However, so far, Linked Open Data does not yet play a significant role among common web technologies that currently facilitate a high-standard Web experience. In this work, we thoroughly discuss the state-of-the-art for Linked Open Data and highlight several shortcomings – some of them we tackle in the main part of this work. First, we propose a novel type of data source meta-information, namely the topics of a dataset. This information could be published with dataset descriptions and support a variety of use cases, such as data source exploration and selection. For the topic retrieval, we present an approach coined Annotated Pattern Percolation (APP), which we evaluate with respect to topics extracted from Wikipedia portals. Second, we contribute to entity linking research by presenting an optimization model for joint entity linking, showing its hardness, and proposing three heuristics implemented in the LINked Data Alignment (LINDA) system. Our first solution can exploit multi-core machines, whereas the second and third approach are designed to run in a distributed shared-nothing environment. We discuss and evaluate the properties of our approaches leading to recommendations which algorithm to use in a specific scenario. The distributed algorithms are among the first of their kind, i.e., approaches for joint entity linking in a distributed fashion. Also, we illustrate that we can tackle the entity linking problem on the very large scale with data comprising more than 100 millions of entity representations from very many sources. Finally, we approach a sub-problem of entity linking, namely the alignment of concepts. We again target a method that looks at the data in its entirety and does not neglect existing relations. Also, this concept alignment method shall execute very fast to serve as a preprocessing for further computations. Our approach, called Holistic Concept Matching (HCM), achieves the required speed through grouping the input by comparing so-called knowledge representations. Within the groups, we perform complex similarity computations, relation conclusions, and detect semantic contradictions. The quality of our result is again evaluated on a large and heterogeneous dataset from the real Web. In summary, this work contributes a set of techniques for enhancing the current state of the Web of Data. All approaches have been tested on large and heterogeneous real-world input.
In this thesis we introduce the concept of the degree of formality. It is directed against a dualistic point of view, which only distinguishes between formal and informal proofs. This dualistic attitude does not respect the differences between the argumentations classified as informal and it is unproductive because the individual potential of the respective argumentation styles cannot be appreciated and remains untapped.
This thesis has two parts. In the first of them we analyse the concept of the degree of formality (including a discussion about the respective benefits for each degree) while in the second we demonstrate its usefulness in three case studies. In the first case study we will repair Haskell B. Curry's view of mathematics, which incidentally is of great importance in the first part of this thesis, in light of the different degrees of formality. In the second case study we delineate how awareness of the different degrees of formality can be used to help students to learn how to prove. Third, we will show how the advantages of proofs of different degrees of formality can be combined by the development of so called tactics having a medium degree of formality. Together the three case studies show that the degrees of formality provide a convincing solution to the problem of untapped potential.
Accurately solving classification problems nowadays is likely to be the most relevant machine learning task. Binary classification separating two classes only is algorithmically simpler but has fewer potential applications as many real-world problems are multi-class. On the reverse, separating only a subset of classes simplifies the classification task. Even though existing multi-class machine learning algorithms are very flexible regarding the number of classes, they assume that the target set Y is fixed and cannot be restricted once the training is finished. On the other hand, existing state-of-the-art production environments are becoming increasingly interconnected with the advance of Industry 4.0 and related technologies such that additional information can simplify the respective classification problems. In light of this, the main aim of this thesis is to introduce dynamic classification that generalizes multi-class classification such that the target class set can be restricted arbitrarily to a non-empty class subset M of Y at any time between two consecutive predictions.
This task is solved by a combination of two algorithmic approaches. First, classifier calibration, which transforms predictions into posterior probability estimates that are intended to be well calibrated. The analysis provided focuses on monotonic calibration and in particular corrects wrong statements that appeared in the literature. It also reveals that bin-based evaluation metrics, which became popular in recent years, are unjustified and should not be used at all. Next, the validity of Platt scaling, which is the most relevant parametric calibration approach, is analyzed in depth. In particular, its optimality for classifier predictions distributed according to four different families of probability distributions as well its equivalence with Beta calibration up to a sigmoidal preprocessing are proven. For non-monotonic calibration, extended variants on kernel density estimation and the ensemble method EKDE are introduced. Finally, the calibration techniques are evaluated using a simulation study with complete information as well as on a selection of 46 real-world data sets.
Building on this, classifier calibration is applied as part of decomposition-based classification that aims to reduce multi-class problems to simpler (usually binary) prediction tasks. For the involved fusing step performed at prediction time, a new approach based on evidence theory is presented that uses classifier calibration to model mass functions. This allows the analysis of decomposition-based classification against a strictly formal background and to prove closed-form equations for the overall combinations. Furthermore, the same formalism leads to a consistent integration of dynamic class information, yielding a theoretically justified and computationally tractable dynamic classification model. The insights gained from this modeling are combined with pairwise coupling, which is one of the most relevant reduction-based classification approaches, such that all individual predictions are combined with a weight. This not only generalizes existing works on pairwise coupling but also enables the integration of dynamic class information.
Lastly, a thorough empirical study is performed that compares all newly introduced approaches to existing state-of-the-art techniques. For this, evaluation metrics for dynamic classification are introduced that depend on corresponding sampling strategies. Thereafter, these are applied during a three-part evaluation. First, support vector machines and random forests are applied on 26 data sets from the UCI Machine Learning Repository. Second, two state-of-the-art deep neural networks are evaluated on five benchmark data sets from a relatively recent reference work. Here, computationally feasible strategies to apply the presented algorithms in combination with large-scale models are particularly relevant because a naive application is computationally intractable. Finally, reference data from a real-world process allowing the inclusion of dynamic class information are collected and evaluated. The results show that in combination with support vector machines and random forests, pairwise coupling approaches yield the best results, while in combination with deep neural networks, differences between the different approaches are mostly small to negligible. Most importantly, all results empirically confirm that dynamic classification succeeds in improving the respective prediction accuracies. Therefore, it is crucial to pass dynamic class information in respective applications, which requires an appropriate digital infrastructure.
As a result of CMOS scaling, radiation-induced Single-Event Effects (SEEs) in electronic circuits became a critical reliability issue for modern Integrated Circuits (ICs) operating under harsh radiation conditions. SEEs can be triggered in combinational or sequential logic by the impact of high-energy particles, leading to destructive or non-destructive faults, resulting in data corruption or even system failure. Typically, the SEE mitigation methods are deployed statically in processing architectures based on the worst-case radiation conditions, which is most of the time unnecessary and results in a resource overhead. Moreover, the space radiation conditions are dynamically changing, especially during Solar Particle Events (SPEs). The intensity of space radiation can differ over five orders of magnitude within a few hours or days, resulting in several orders of magnitude fault probability variation in ICs during SPEs. This thesis introduces a comprehensive approach for designing a self-adaptive fault resilient multiprocessing system to overcome the static mitigation overhead issue. This work mainly addresses the following topics: (1) Design of on-chip radiation particle monitor for real-time radiation environment detection, (2) Investigation of space environment predictor, as support for solar particle events forecast, (3) Dynamic mode configuration in the resilient multiprocessing system. Therefore, according to detected and predicted in-flight space radiation conditions, the target system can be configured to use no mitigation or low-overhead mitigation during non-critical periods of time. The redundant resources can be used to improve system performance or save power. On the other hand, during increased radiation activity periods, such as SPEs, the mitigation methods can be dynamically configured appropriately depending on the real-time space radiation environment, resulting in higher system reliability. Thus, a dynamic trade-off in the target system between reliability, performance and power consumption in real-time can be achieved. All results of this work are evaluated in a highly reliable quad-core multiprocessing system that allows the self-adaptive setting of optimal radiation mitigation mechanisms during run-time. Proposed methods can serve as a basis for establishing a comprehensive self-adaptive resilient system design process. Successful implementation of the proposed design in the quad-core multiprocessor shows its application perspective also in the other designs.
Contemporary multi-core processors are parallel systems that also provide shared memory for programs running on them. Both the increasing number of cores in so-called many-core systems and the still growing computational power of the cores demand for memory systems that are able to deliver high bandwidths. Caches are essential components to satisfy this requirement. Nevertheless, hardware-based cache coherence in many-core chips faces practical limits to provide both coherence and high memory bandwidths. In addition, a shift away from global coherence can be observed. As a result, alternative architectures and suitable programming models need to be investigated.
This thesis focuses on fast communication for non-cache-coherent many-core architectures. Experiments are conducted on the Single-Chip Cloud Computer (SCC), a non-cache-coherent many-core processor with 48 mesh-connected cores. Although originally designed for message passing, the results of this thesis show that shared memory can be efficiently used for one-sided communication on this kind of architecture. One-sided communication enables data exchanges between processes where the receiver is not required to know the details of the performed communication. In the notion of the Message Passing Interface (MPI) standard, this type of communication allows to access memory of remote processes. In order to support this communication scheme on non-cache-coherent architectures, both an efficient process synchronization and a communication scheme with software-managed cache coherence are designed and investigated.
The process synchronization realizes the concept of the general active target synchronization scheme from the MPI standard. An existing classification of implementation approaches is extended and used to identify an appropriate class for the non-cache-coherent shared memory platform. Based on this classification, existing implementations are surveyed in order to find beneficial concepts, which are then used to design a lightweight synchronization protocol for the SCC that uses shared memory and uncached memory accesses. The proposed scheme is not prone to process skew and also enables direct communication as soon as both communication partners are ready. Experimental results show very good scaling properties and up to five times lower synchronization latency compared to a tuned message-based MPI implementation for the SCC.
For the communication, SCOSCo, a shared memory approach with software-managed cache coherence, is presented. According requirements for the coherence that fulfill MPI's separate memory model are formulated, and a lightweight implementation exploiting SCC hard- and software features is developed. Despite a discovered malfunction in the SCC's memory subsystem, the experimental evaluation of the design reveals up to five times better bandwidths and nearly four times lower latencies in micro-benchmarks compared to the SCC-tuned but message-based MPI library. For application benchmarks, like a parallel 3D fast Fourier transform, the runtime share of communication can be reduced by a factor of up to five. In addition, this thesis postulates beneficial hardware concepts that would support software-managed coherence for one-sided communication on future non-cache-coherent architectures where coherence might be only available in local subdomains but not on a global processor level.
Cloud computing is a model for enabling on-demand access to a shared pool of computing resources. With virtually limitless on-demand resources, a cloud environment enables the hosted Internet application to quickly cope when there is an increase in the workload. However, the overhead of provisioning resources exposes the Internet application to periods of under-provisioning and performance degradation. Moreover, the performance interference, due to the consolidation in the cloud environment, complicates the performance management of the Internet applications. In this dissertation, we propose two approaches to mitigate the impact of the resources provisioning overhead. The first approach employs control theory to scale resources vertically and cope fast with workload. This approach assumes that the provider has knowledge and control over the platform running in the virtual machines (VMs), which limits it to Platform as a Service (PaaS) and Software as a Service (SaaS) providers. The second approach is a customer-side one that deals with the horizontal scalability in an Infrastructure as a Service (IaaS) model. It addresses the trade-off problem between cost and performance with a multi-goal optimization solution. This approach finds the scale thresholds that achieve the highest performance with the lowest increase in the cost. Moreover, the second approach employs a proposed time series forecasting algorithm to scale the application proactively and avoid under-utilization periods. Furthermore, to mitigate the interference impact on the Internet application performance, we developed a system which finds and eliminates the VMs suffering from performance interference. The developed system is a light-weight solution which does not imply provider involvement. To evaluate our approaches and the designed algorithms at large-scale level, we developed a simulator called (ScaleSim). In the simulator, we implemented scalability components acting as the scalability components of Amazon EC2. The current scalability implementation in Amazon EC2 is used as a reference point for evaluating the improvement in the scalable application performance. ScaleSim is fed with realistic models of the RUBiS benchmark extracted from the real environment. The workload is generated from the access logs of the 1998 world cup website. The results show that optimizing the scalability thresholds and adopting proactive scalability can mitigate 88% of the resources provisioning overhead impact with only a 9% increase in the cost.
With the rise of electronic integration between organizations, the need for a precise specification of interaction behavior increases. Information systems, replacing interaction previously carried out by humans via phone, faxes and emails, require a precise specification for handling all possible situations. Such interaction behavior is described in process choreographies. Choreographies enumerate the roles involved, the allowed interactions, the message contents and the behavioral dependencies between interactions. Choreographies serve as interaction contract and are the starting point for adapting existing business processes and systems or for implementing new software components. As a thorough analysis and comparison of choreography modeling languages is missing in the literature, this thesis introduces a requirements framework for choreography languages and uses it for comparing current choreography languages. Language proposals for overcoming the limitations are given for choreography modeling on the conceptual and on the technical level. Using an interconnection modeling style, behavioral dependencies are defined on a per-role basis and different roles are interconnected using message flow. This thesis reveals a number of modeling "anti-patterns" for interconnection modeling, motivating further investigations on choreography languages following the interaction modeling style. Here, interactions are seen as atomic building blocks and the behavioral dependencies between them are defined globally. Two novel language proposals are put forward for this modeling style which have already influenced industrial standardization initiatives. While avoiding many of the pitfalls of interconnection modeling, new anomalies can arise in interaction models. A choreography might not be realizable, i.e. there does not exist a set of interacting roles that collectively realize the specified behavior. This thesis investigates different dimensions of realizability.
Services that operate over the Internet are under constant threat of being exposed to fraudulent use. Maintaining good user experience for legitimate users often requires the classification of entities as malicious or legitimate in order to initiate countermeasures. As an example, inbound email spam filters decide for spam or non-spam. They can base their decision on both the content of each email as well as on features that summarize prior emails received from the sending server. In general, discriminative classification methods learn to distinguish positive from negative entities. Each decision for a label may be based on features of the entity and related entities. When labels of related entities have strong interdependencies---as can be assumed e.g. for emails being delivered by the same user---classification decisions should not be made independently and dependencies should be modeled in the decision function. This thesis addresses the formulation of discriminative classification problems that are tailored for the specific demands of the following three Internet security applications. Theoretical and algorithmic solutions are devised to protect an email service against flooding of user inboxes, to mitigate abusive usage of outbound email servers, and to protect web servers against distributed denial of service attacks.
In the application of filtering an inbound email stream for unsolicited emails, utilizing features that go beyond each individual email's content can be valuable. Information about each sending mail server can be aggregated over time and may help in identifying unwanted emails. However, while this information will be available to the deployed email filter, some parts of the training data that are compiled by third party providers may not contain this information. The missing features have to be estimated at training time in order to learn a classification model. In this thesis an algorithm is derived that learns a decision function that integrates over a distribution of values for each missing entry. The distribution of missing values is a free parameter that is optimized to learn an optimal decision function.
The outbound stream of emails of an email service provider can be separated by the customer IDs that ask for delivery. All emails that are sent by the same ID in the same period of time are related, both in content and in label. Hijacked customer accounts may send batches of unsolicited emails to other email providers, which in turn might blacklist the sender's email servers after detection of incoming spam emails. The risk of being blocked from further delivery depends on the rate of outgoing unwanted emails and the duration of high spam sending rates. An optimization problem is developed that minimizes the expected cost for the email provider by learning a decision function that assigns a limit on the sending rate to customers based on the each customer's email stream.
Identifying attacking IPs during HTTP-level DDoS attacks allows to block those IPs from further accessing the web servers. DDoS attacks are usually carried out by infected clients that are members of the same botnet and show similar traffic patterns. HTTP-level attacks aim at exhausting one or more resources of the web server infrastructure, such as CPU time. If the joint set of attackers cannot increase resource usage close to the maximum capacity, no effect will be experienced by legitimate users of hosted web sites. However, if the additional load raises the computational burden towards the critical range, user experience will degrade until service may be unavailable altogether. As the loss of missing one attacker depends on block decisions for other attackers---if most other attackers are detected, not blocking one client will likely not be harmful---a structured output model has to be learned. In this thesis an algorithm is developed that learns a structured prediction decoder that searches the space of label assignments, guided by a policy.
Each model is evaluated on real-world data and is compared to reference methods. The results show that modeling each classification problem according to the specific demands of the task improves performance over solutions that do not consider the constraints inherent to an application.
The goal of a Brain-Computer Interface (BCI) consists of the development of a unidirectional interface between a human and a computer to allow control of a device only via brain signals. While the BCI systems of almost all other groups require the user to be trained over several weeks or even months, the group of Prof. Dr. Klaus-Robert Müller in Berlin and Potsdam, which I belong to, was one of the first research groups in this field which used machine learning techniques on a large scale. The adaptivity of the processing system to the individual brain patterns of the subject confers huge advantages for the user. Thus BCI research is considered a hot topic in machine learning and computer science. It requires interdisciplinary cooperation between disparate fields such as neuroscience, since only by combining machine learning and signal processing techniques based on neurophysiological knowledge will the largest progress be made. In this work I particularly deal with my part of this project, which lies mainly in the area of computer science. I have considered the following three main points: <b>Establishing a performance measure based on information theory:</b> I have critically illuminated the assumptions of Shannon's information transfer rate for application in a BCI context. By establishing suitable coding strategies I was able to show that this theoretical measure approximates quite well to what is practically achieveable. <b>Transfer and development of suitable signal processing and machine learning techniques:</b> One substantial component of my work was to develop several machine learning and signal processing algorithms to improve the efficiency of a BCI. Based on the neurophysiological knowledge that several independent EEG features can be observed for some mental states, I have developed a method for combining different and maybe independent features which improved performance. In some cases the performance of the combination algorithm outperforms the best single performance by more than 50 %. Furthermore, I have theoretically and practically addressed via the development of suitable algorithms the question of the optimal number of classes which should be used for a BCI. It transpired that with BCI performances reported so far, three or four different mental states are optimal. For another extension I have combined ideas from signal processing with those of machine learning since a high gain can be achieved if the temporal filtering, i.e., the choice of frequency bands, is automatically adapted to each subject individually. <b>Implementation of the Berlin brain computer interface and realization of suitable experiments:</b> Finally a further substantial component of my work was to realize an online BCI system which includes the developed methods, but is also flexible enough to allow the simple realization of new algorithms and ideas. So far, bitrates of up to 40 bits per minute have been achieved with this system by absolutely untrained users which, compared to results of other groups, is highly successful.
In order to face the rapidly increasing need for computational resources of various scientific and engineering applications one has to think of new ways to make more efficient use of the worlds current computational resources. In this respect, the growing speed of wide area networks made a new kind of distributed computing possible: Metacomputing or (distributed) Grid computing. This is a rather new and uncharted field in computational science. The rapidly increasing speed of networks even outperforms the average increase of processor speed: Processor speeds double on average each 18 month whereas network bandwidths double every 9 months. Due to this development of local and wide area networks Grid computing will certainly play a key role in the future of parallel computing. This type of distributed computing, however, distinguishes from the traditional parallel computing in many ways since it has to deal with many problems not occurring in classical parallel computing. Those problems are for example heterogeneity, authentication and slow networks to mention only a few. Some of those problems, e.g. the allocation of distributed resources along with the providing of information about these resources to the application have been already attacked by the Globus software. Unfortunately, as far as we know, hardly any application or middle-ware software takes advantage of this information, since most parallelizing algorithms for finite differencing codes are implicitly designed for single supercomputer or cluster execution. We show that although it is possible to apply classical parallelizing algorithms in a Grid environment, in most cases the observed efficiency of the executed code is very poor. In this work we are closing this gap. In our thesis, we will - show that an execution of classical parallel codes in Grid environments is possible but very slow - analyze this situation of bad performance, nail down bottlenecks in communication, remove unnecessary overhead and other reasons for low performance - develop new and advanced algorithms for parallelisation that are aware of a Grid environment in order to generelize the traditional parallelization schemes - implement and test these new methods, replace and compare with the classical ones - introduce dynamic strategies that automatically adapt the running code to the nature of the underlying Grid environment. The higher the performance one can achieve for a single application by manual tuning for a Grid environment, the lower the chance that those changes are widely applicable to other programs. In our analysis as well as in our implementation we tried to keep the balance between high performance and generality. None of our changes directly affect code on the application level which makes our algorithms applicable to a whole class of real world applications. The implementation of our work is done within the Cactus framework using the Globus toolkit, since we think that these are the most reliable and advanced programming frameworks for supporting computations in Grid environments. On the other hand, however, we tried to be as general as possible, i.e. all methods and algorithms discussed in this thesis are independent of Cactus or Globus.
Modern biological analysis techniques supply scientists with various forms of data. One category of such data are the so called "expression data". These data indicate the quantities of biochemical compounds present in tissue samples. Recently, expression data can be generated at a high speed. This leads in turn to amounts of data no longer analysable by classical statistical techniques. Systems biology is the new field that focuses on the modelling of this information. At present, various methods are used for this purpose. One superordinate class of these methods is machine learning. Methods of this kind had, until recently, predominantly been used for classification and prediction tasks. This neglected a powerful secondary benefit: the ability to induce interpretable models. Obtaining such models from data has become a key issue within Systems biology. Numerous approaches have been proposed and intensively discussed. This thesis focuses on the examination and exploitation of one basic technique: decision trees. The concept of comparing sets of decision trees is developed. This method offers the possibility of identifying significant thresholds in continuous or discrete valued attributes through their corresponding set of decision trees. Finding significant thresholds in attributes is a means of identifying states in living organisms. Knowing about states is an invaluable clue to the understanding of dynamic processes in organisms. Applied to metabolite concentration data, the proposed method was able to identify states which were not found with conventional techniques for threshold extraction. A second approach exploits the structure of sets of decision trees for the discovery of combinatorial dependencies between attributes. Previous work on this issue has focused either on expensive computational methods or the interpretation of single decision trees a very limited exploitation of the data. This has led to incomplete or unstable results. That is why a new method is developed that uses sets of decision trees to overcome these limitations. Both the introduced methods are available as software tools. They can be applied consecutively or separately. That way they make up a package of analytical tools that usefully supplement existing methods. By means of these tools, the newly introduced methods were able to confirm existing knowledge and to suggest interesting and new relationships between metabolites.
This thesis presents an attempt to use source code synthesised from Coq formalisations of device drivers for existing (micro)kernel operating systems, with a particular focus on the Linux Kernel.
In the first part, the technical background and related work are described. The focus is here on the possible approaches to synthesising certified software with Coq, namely the extraction to functional languages using the Coq extraction plugin and the extraction to Clight code using the CertiCoq plugin. It is noted that the implementation of CertiCoq is verified, whereas this is not the case for the Coq extraction plugin. Consequently, there is a correctness guarantee for the generated Clight code which does not hold for the code being generated by the Coq extraction plugin. Furthermore, the differences between user space and kernel space software are discussed in relation to Linux device drivers. It is elaborated that it is not possible to generate working Linux kernel module components using the Coq extraction plugin without significant modifications. In contrast, it is possible to produce working user space drivers both with the Coq extraction plugin and CertiCoq. The subsequent parts describe the main contributions of the thesis.
In the second part, it is demonstrated how to extend the Coq extraction plugin to synthesise foreign function calls between the functional language OCaml and the imperative language C. This approach has the potential to improve the type-safety of user space drivers. Furthermore, it is shown that the code being synthesised by CertiCoq cannot be used in kernel space without modifications to the necessary runtime. Consequently, the necessary modifications to the runtimes of CertiCoq and VeriFFI are introduced, resulting in the runtimes becoming compatible components of a Linux kernel module. Furthermore, justifications for the transformations are provided and possible further extensions to both plugins and solutions to failing garbage collection calls in kernel space are discussed.
The third part presents a proof of concept device driver for the Linux Kernel. To achieve this, the event handler of the original PC Speaker driver is partially formalised in Coq. Furthermore, some relevant formal properties of the formalised functionality are discussed. Subsequently, a kernel module is defined, utilising the modified variants of CertiCoq and VeriFFI to compile a working device driver. It is furthermore shown that it is possible to compile the synthesised code with CompCert, thereby extending the guarantee of correctness to the assembly layer. This is followed by a performance evaluation that compares a naive formalisation of the PC speaker functionality with the original PC Speaker driver pointing out the weaknesses in the formalisation and possible improvements. The part closes with a summary of the results, their implications and open questions being raised.
The last part lists all used sources, separated into scientific literature, documentations or reference manuals and artifacts, i.e. source code.
Answer Set Programming (ASP) is an emerging paradigm for declarative programming, in which a computational problem is specified by a logic program such that particular models, called answer sets, match solutions. ASP faces a growing range of applications, demanding for high-performance tools able to solve complex problems. ASP integrates ideas from a variety of neighboring fields. In particular, automated techniques to search for answer sets are inspired by Boolean Satisfiability (SAT) solving approaches. While the latter have firm proof-theoretic foundations, ASP lacks formal frameworks for characterizing and comparing solving methods. Furthermore, sophisticated search patterns of modern SAT solvers, successfully applied in areas like, e.g., model checking and verification, are not yet established in ASP solving. We address these deficiencies by, for one, providing proof-theoretic frameworks that allow for characterizing, comparing, and analyzing approaches to answer set computation. For another, we devise modern ASP solving algorithms that integrate and extend state-of-the-art techniques for Boolean constraint solving. We thus contribute to the understanding of existing ASP solving approaches and their interconnections as well as to their enhancement by incorporating sophisticated search patterns. The central idea of our approach is to identify atomic as well as composite constituents of a propositional logic program with Boolean variables. This enables us to describe fundamental inference steps, and to selectively combine them in proof-theoretic characterizations of various ASP solving methods. In particular, we show that different concepts of case analyses applied by existing ASP solvers implicate mutual exponential separations regarding their best-case complexities. We also develop a generic proof-theoretic framework amenable to language extensions, and we point out that exponential separations can likewise be obtained due to case analyses on them. We further exploit fundamental inference steps to derive Boolean constraints characterizing answer sets. They enable the conception of ASP solving algorithms including search patterns of modern SAT solvers, while also allowing for direct technology transfers between the areas of ASP and SAT solving. Beyond the search for one answer set of a logic program, we address the enumeration of answer sets and their projections to a subvocabulary, respectively. The algorithms we develop enable repetition-free enumeration in polynomial space without being intrusive, i.e., they do not necessitate any modifications of computations before an answer set is found. Our approach to ASP solving is implemented in clasp, a state-of-the-art Boolean constraint solver that has successfully participated in recent solver competitions. Although we do here not address the implementation techniques of clasp or all of its features, we present the principles of its success in the context of ASP solving.
Quantified Boolean formulas (QBFs) play an important role in theoretical computer science. QBF extends propositional logic in such a way that many advanced forms of reasoning can be easily formulated and evaluated. In this dissertation we present our ZQSAT, which is an algorithm for evaluating quantified Boolean formulas. ZQSAT is based on ZBDD: Zero-Suppressed Binary Decision Diagram , which is a variant of BDD, and an adopted version of the DPLL algorithm. It has been implemented in C using the CUDD: Colorado University Decision Diagram package. The capability of ZBDDs in storing sets of subsets efficiently enabled us to store the clauses of a QBF very compactly and let us to embed the notion of memoization to the DPLL algorithm. These points led us to implement the search algorithm in such a way that we could store and reuse the results of all previously solved subformulas with a little overheads. ZQSAT can solve some sets of standard QBF benchmark problems (known to be hard for DPLL based algorithms) faster than the best existing solvers. In addition to prenex-CNF, ZQSAT accepts prenex-NNF formulas. We show and prove how this capability can be exponentially beneficial.
Virtual 3D city and landscape models are the main subject investigated in this thesis. They digitally represent urban space and have many applications in different domains, e.g., simulation, cadastral management, and city planning. Visualization is an elementary component of these applications. Photo-realistic visualization with an increasingly high degree of detail leads to fundamental problems for comprehensible visualization. A large number of highly detailed and textured objects within a virtual 3D city model may create visual noise and overload the users with information. Objects are subject to perspective foreshortening and may be occluded or not displayed in a meaningful way, as they are too small. In this thesis we present abstraction techniques that automatically process virtual 3D city and landscape models to derive abstracted representations. These have a reduced degree of detail, while essential characteristics are preserved. After introducing definitions for model, scale, and multi-scale representations, we discuss the fundamentals of map generalization as well as techniques for 3D generalization. The first presented technique is a cell-based generalization of virtual 3D city models. It creates abstract representations that have a highly reduced level of detail while maintaining essential structures, e.g., the infrastructure network, landmark buildings, and free spaces. The technique automatically partitions the input virtual 3D city model into cells based on the infrastructure network. The single building models contained in each cell are aggregated to abstracted cell blocks. Using weighted infrastructure elements, cell blocks can be computed on different hierarchical levels, storing the hierarchy relation between the cell blocks. Furthermore, we identify initial landmark buildings within a cell by comparing the properties of individual buildings with the aggregated properties of the cell. For each block, the identified landmark building models are subtracted using Boolean operations and integrated in a photo-realistic way. Finally, for the interactive 3D visualization we discuss the creation of the virtual 3D geometry and their appearance styling through colors, labeling, and transparency. We demonstrate the technique with example data sets. Additionally, we discuss applications of generalization lenses and transitions between abstract representations. The second technique is a real-time-rendering technique for geometric enhancement of landmark objects within a virtual 3D city model. Depending on the virtual camera distance, landmark objects are scaled to ensure their visibility within a specific distance interval while deforming their environment. First, in a preprocessing step a landmark hierarchy is computed, this is then used to derive distance intervals for the interactive rendering. At runtime, using the virtual camera distance, a scaling factor is computed and applied to each landmark. The scaling factor is interpolated smoothly at the interval boundaries using cubic Bézier splines. Non-landmark geometry that is near landmark objects is deformed with respect to a limited number of landmarks. We demonstrate the technique by applying it to a highly detailed virtual 3D city model and a generalized 3D city model. In addition we discuss an adaptation of the technique for non-linear projections and mobile devices. The third technique is a real-time rendering technique to create abstract 3D isocontour visualization of virtual 3D terrain models. The virtual 3D terrain model is visualized as a layered or stepped relief. The technique works without preprocessing and, as it is implemented using programmable graphics hardware, can be integrated with minimal changes into common terrain rendering techniques. Consequently, the computation is done in the rendering pipeline for each vertex, primitive, i.e., triangle, and fragment. For each vertex, the height is quantized to the nearest isovalue. For each triangle, the vertex configuration with respect to their isovalues is determined first. Using the configuration, the triangle is then subdivided. The subdivision forms a partial step geometry aligned with the triangle. For each fragment, the surface appearance is determined, e.g., depending on the surface texture, shading, and height-color-mapping. Flexible usage of the technique is demonstrated with applications from focus+context visualization, out-of-core terrain rendering, and information visualization. This thesis presents components for the creation of abstract representations of virtual 3D city and landscape models. Re-using visual language from cartography, the techniques enable users to build on their experience with maps when interpreting these representations. Simultaneously, characteristics of 3D geovirtual environments are taken into account by addressing and discussing, e.g., continuous scale, interaction, and perspective.
The business problem of having inefficient processes, imprecise process analyses, and simulations as well as non-transparent artificial neuronal network models can be overcome by an easy-to-use modeling concept. With the aim of developing a flexible and efficient approach to modeling, simulating, and optimizing processes, this paper proposes a flexible Concept of Neuronal Modeling (CoNM). The modeling concept, which is described by the modeling language designed and its mathematical formulation and is connected to a technical substantiation, is based on a collection of novel sub-artifacts. As these have been implemented as a computational model, the set of CoNM tools carries out novel kinds of Neuronal Process Modeling (NPM), Neuronal Process Simulations (NPS), and Neuronal Process Optimizations (NPO). The efficacy of the designed artifacts was demonstrated rigorously by means of six experiments and a simulator of real industrial production processes.
Learning a model for the relationship between the attributes and the annotated labels of data examples serves two purposes. Firstly, it enables the prediction of the label for examples without annotation. Secondly, the parameters of the model can provide useful insights into the structure of the data. If the data has an inherent partitioned structure, it is natural to mirror this structure in the model. Such mixture models predict by combining the individual predictions generated by the mixture components which correspond to the partitions in the data. Often the partitioned structure is latent, and has to be inferred when learning the mixture model. Directly evaluating the accuracy of the inferred partition structure is, in many cases, impossible because the ground truth cannot be obtained for comparison. However it can be assessed indirectly by measuring the prediction accuracy of the mixture model that arises from it. This thesis addresses the interplay between the improvement of predictive accuracy by uncovering latent cluster structure in data, and further addresses the validation of the estimated structure by measuring the accuracy of the resulting predictive model. In the application of filtering unsolicited emails, the emails in the training set are latently clustered into advertisement campaigns. Uncovering this latent structure allows filtering of future emails with very low false positive rates. In order to model the cluster structure, a Bayesian clustering model for dependent binary features is developed in this thesis. Knowing the clustering of emails into campaigns can also aid in uncovering which emails have been sent on behalf of the same network of captured hosts, so-called botnets. This association of emails to networks is another layer of latent clustering. Uncovering this latent structure allows service providers to further increase the accuracy of email filtering and to effectively defend against distributed denial-of-service attacks. To this end, a discriminative clustering model is derived in this thesis that is based on the graph of observed emails. The partitionings inferred using this model are evaluated through their capacity to predict the campaigns of new emails. Furthermore, when classifying the content of emails, statistical information about the sending server can be valuable. Learning a model that is able to make use of it requires training data that includes server statistics. In order to also use training data where the server statistics are missing, a model that is a mixture over potentially all substitutions thereof is developed. Another application is to predict the navigation behavior of the users of a website. Here, there is no a priori partitioning of the users into clusters, but to understand different usage scenarios and design different layouts for them, imposing a partitioning is necessary. The presented approach simultaneously optimizes the discriminative as well as the predictive power of the clusters. Each model is evaluated on real-world data and compared to baseline methods. The results show that explicitly modeling the assumptions about the latent cluster structure leads to improved predictions compared to the baselines. It is beneficial to incorporate a small number of hyperparameters that can be tuned to yield the best predictions in cases where the prediction accuracy can not be optimized directly.
Independent component analysis (ICA) is a tool for statistical data analysis and signal processing that is able to decompose multivariate signals into their underlying source components. Although the classical ICA model is highly useful, there are many real-world applications that require powerful extensions of ICA. This thesis presents new methods that extend the functionality of ICA: (1) reliability and grouping of independent components with noise injection, (2) robust and overcomplete ICA with inlier detection, and (3) nonlinear ICA with kernel methods.
In the last decades, there was a notable progress in solving the well-known Boolean satisfiability (Sat) problem, which can be witnessed by powerful Sat solvers. One of the reasons why these solvers are so fast are structural properties of instances that are utilized by the solver’s interna. This thesis deals with the well-studied structural property treewidth, which measures the closeness of an instance to being a tree. In fact, there are many problems parameterized by treewidth that are solvable in polynomial time in the instance size when parameterized by treewidth.
In this work, we study advanced treewidth-based methods and tools for problems in knowledge representation and reasoning (KR). Thereby, we provide means to establish precise runtime results (upper bounds) for canonical problems relevant to KR. Then, we present a new type of problem reduction, which we call decomposition-guided (DG) that
allows us to precisely monitor the treewidth when reducing from one problem to another problem. This new reduction type will be the basis for a long-open lower bound result for quantified Boolean formulas and allows us to design a new methodology for establishing runtime lower bounds for problems parameterized by treewidth.
Finally, despite these lower bounds, we provide an efficient implementation of algorithms that adhere to treewidth. Our approach finds suitable abstractions of instances, which are subsequently refined in a recursive fashion, and it uses Sat solvers for solving subproblems. It turns out that our resulting solver is quite competitive for two canonical counting problems related to Sat.