<p>This thesis gives a partial answer to the question: “Can type systems detect modeling errors in scientific computing, particularly for inverse problems derived from physical models?” by considering, in detail, the major aspects of inverse problems in Magnetic Resonance Imaging (MRI). We define a type-system that can capture all correctness properties for MRI inverse problems, including many properties that are not captured with current type-systems, e.g., frames of reference. We implemented a type-system in the Haskell language that can capture the errors arising in translating a mathe- matical model into a linear or nonlinear system, or alternatively into an objective function. Most models are (or can be approximated by) linear transformations, and we demonstrate the feasibility of capturing their correctness at the type level using what is arguably the most difficult case, the (discrete) Fourier transformation (DFT). By this, we mean that we are able to catch, at compile time, all known errors in ap- plying the DFT. The first part of this thesis describes the Haskell implementation of vector size, physical units, frame of reference, and so on required in the mathemat- ical modelling of inverse problems without regularization. To practically solve most inverse problems, especially those including noisy data or ill-conditioned systems, one must use regularization. The second part of this thesis addresses the question of defining new regularizers and identifying existing regularizers the correctness of which (in our estimation) can be formally verified at the type level. We describe such Bayesian regularization schemes based on probability theory, and describe a novel simple regularizer of this type. We leave as future work the formalization of such regularizers.</p> / Master of Science (MSc)

<p>The purpose of this research has been to reassess the Ag-Mg system using the CALPHAD technique. Compared with previous assessments, we carry out the optimization by fitting calculations to the original data instead of second-hand information. Moreover, we use a two sub-lattice model and a four sub-lattice model based on compound energy formalism to simulate both first-order and second-order transformations between the FCC phase and the L1₂ phase. Undoubtedly, the CALPHAD technique has achieved a degree of maturity, but its deficiencies are regularly ignored.</p> <p>In this thesis, we develop an interval method based on Kantorovich’s idea to overcome the shortcomings of the CALPHAD technique. Both advantages and disadvantages of the interval method are discussed. We also present an example of the interval approach on thermodynamic optimization of the Ag-Mg melt. The results suggest that this method would be helpful as a pre-optimization tool.</p> / Master of Applied Science (MASc)

<p dir="ltr">Heterogeneous clusters with nodes containing one or more accelerators, such as GPUs, have become common. While MPI provides inter-address space communication, and OpenCL provides a process with access to heterogeneous computational resources, programmers are forced to write hybrid programs that manage the interaction of both of these systems. This paper describes an array programming interface that provides users with automatic and manual distributions of data and work. Using work distribution and kernel def and use information, communication among processes and devices in a process is performed automatically. By providing a unified programming model to the user, program development is simplified.</p>

<p dir="ltr">Protecting personal information is growing increasingly important to the general public, to the point that major tech companies now advertise the privacy features of their products. Despite this, it remains challenging to implement applications that do not leak private information either directly or indirectly, through timing behavior, memory access patterns, or control flow side channels. Existing security and cryptographic techniques such as secure multiparty computation (MPC) provide solutions to privacy-preserving computation, but they can be difficult to use for non-experts and even experts.</p><p dir="ltr">This dissertation develops the design, theory and implementation of various language-based techniques that help programmers write privacy-critical applications under a strong threat model. The proposed languages support private structured data, such as trees, that may hide their structural information and complex policies that go beyond whether a particular field of a record is private. More crucially, the approaches described in this dissertation decouple privacy and programmatic concerns, allowing programmers to implement privacy-preserving applications modularly, i.e., to independently develop application logic and independently update and audit privacy policies. Secure-by-construction applications are derived automatically by combining a standard program with a separately specified security policy.</p><p><br></p>

Cette thèse présente un algorithme qui élimine les racines carrées et les divi- sions dans des programmes sans boucles, utilisés dans des systèmes embarqués, tout en préservant la sémantique. L'élimination de ces opérations permet d'éviter les erreurs d'arrondis à l'exécution, ces erreurs d'arrondis pouvant entraîner un comportement complètement inattendu de la part du programme. Cette trans- formation respecte les contraintes du code embarqué, en particulier la nécessité pour le programme produit de s'exécuter en mémoire fixe. Cette transformation utilise deux algorithmes fondamentaux développés dans cette thèse. Le premier permet d'éliminer les racines carrées et les divisions des expressions booléennes contenant des comparaisons d'expressions arithmétiques. Le second est un algo- rithme qui résout un problème d'anti-unification particulier, que nous appelons anti-unification contrainte. Cette transformation de programme est définie et prou- vée dans l'assistant de preuves PVS. Elle est aussi implantée comme une stratégie de ce système. L'anti-unification contrainte est aussi utilisée pour étendre la transformation à des programmes contenant des fonctions. Elle permet ainsi d'éliminer les racines carrées et les divisions de spécifications écrites en PVS. La robustesse de cette méthode est mise en valeur par un exemple conséquent: l'élimination des racines carrées et des divisions dans un programme de détection des conflits aériens.

Indiana University-Purdue University Indianapolis (IUPUI) / The contribution of this thesis focuses on addressing the challenges of improving and integrating the UniFrame Discovery Service (URDS) and Multi-level Matching (MLM) concepts. The objective was to find enhancements for both URDS and MLM and address the need of a comprehensive discovery service which goes beyond simple attribute based matching. It presents a detailed discussion on developing an enhanced version of URDS with MLM (proURDS). After implementing proURDS, the thesis includes details of experiments with different deployments of URDS components and different configurations of MLM. The experiments and analysis were carried out using proURDS produced MLM contracts. The proURDS referred to a public dataset called QWS dataset. This dataset includes actual information of software components (i.e., web services), which were harvested from the Internet. The proURDS implements the different matching operations as independent operators at each level of matching (i.e., General, Syntactic, Semantic, Synchronization, and QoS). Finally, a case study was carried out with the deployed proURDS. The case study addresses real world component discovery requirements from the earth science domain. It uses the contracts collected from public portals which provide geographical and weather related data.

This MSc Thesis is focused on the multimodeling area, especially on the cooperation of the C++ language and the Prolog language. The recent research is established on my semester study, which also dealt with the multimodeling area. During this research a prototype of interconnection library for cooperation of above mentioned programming languages was developed. This prototype of the library was finished within the scope of this thesis and some case-studies were created, too, using also another simulation library - SIMLIB/C++. These case-studies have their focus in the problems of artificial intelligence. The main benefit of this thesis is the confrontation of methods with and without using artificial intelligence.

Blind and visually impaired students are under-represented in the science, technology, engineering, and mathematics disciplines of higher education and the workforce. This is due primarily to the difficulties they encounter in trying to succeed in mathematics courses. While there are sufficient tools available to create Braille content, including the special Nemeth Braille used in the U.S. for mathematics constructs, there are very few tools to allow a blind or visually impaired student to create his/her own mathematical content in a manner that sighted individuals can use. The software tools that are available are isolated, do not interface well with other common software, and may be priced for institutional use instead of individual use. Instructors are unprepared or unable to interact with these students in a real-time manner. All of these factors combine to isolate the blind or visually impaired student in the study of mathematics. Nemeth Braille is a complete mathematical markup system in Braille, containing everything that is needed to produce quality math content at all levels of complexity. Blind and visually impaired students should not have to learn any additional markup languages in order to produce math content. This work addressed the needs of the individual blind or visually impaired student who must be able to produce mathematical content for course assignments, and who wishes to interact with peers and instructors on a real-time basis to share mathematical content. Two tools were created to facilitate mathematical interaction: a Nemeth Braille editor, and a real-time instant messenger chat capability that supports Nemeth Braille and MathML constructs. In the Visually Impaired view, the editor accepts Nemeth Braille input, displays the math expressions in a tree structure which will allow sub-expressions to be expanded or collapsed. The Braille constructs can be translated to MathML for display within MathType. Similarly, in the Sighted view, math constructs entered in MathType can be translated into Nemeth Braille. Mathematical content can then be shared between sighted and visually impaired users via the instant messenger chat capability. Using Math in the Dark software, blind and visually impaired students can work math problems fully in Nemeth Braille and can seamlessly convert their work into MathML for viewing by sighted instructors. The converted output has the quality of professionally produced math content. Blind and VI students can also communicate and share math constructs with a sighted partner via a real-time chat feature, with automatic translation in both directions, allowing VI students to obtain help in real-time from a sighted instructor or tutor. By eliminating the burden of translation, this software will help to remove the barriers faced by blind and VI students who wish to excel in the STEM fields of study.

Predictive statistical modeling shows promise in accurately predicting academic performance for students enrolled in online programs. This approach has proven effective in accurately identifying students who are at-risk enabling instructors to provide instructional intervention. While the potential benefits of statistical modeling is significant, implementations have proven to be complex, costly, and difficult to maintain. To address these issues, the purpose of this study is to develop a fully integrated, automated predictive modeling system (PMS) that is flexible, easy to use, and portable to identify students who are potentially at-risk for not succeeding in a course they are currently enrolled in. Dynamic and static variables from a student system (edX) will be analyzed to predict academic performance of an individual student or entire class. The PMS model framework will include development of an open-source Web application, application programming interface (API), and SQL reporting services (SSRS). The model is based on knowledge discovery database (KDD) approach utilizing inductive logic programming language (ILP) to analyze student data. This alternative approach for predicting academic performance has several unique advantages over current predictive modeling techniques in use and is a promising new direction in educational research.

Les logiciels modernes sont de plus en plus complexes. Ceci est dû en partie à l'hétérogénéité des solutions misent en oeuvre, au caractère distribué des architectures de déploiement et à la dynamicité requise pour de tels logiciels qui devraient être capable de s'adapter en fonction des variations de leur contexte d'évolution. D'un autre coté, l'importance grandissante des contraintes de productivité dans le but de réduire les coûts de maintenance et de production des systèmes informatiques a favorisé l'émergence de nouveaux paradigmes pour répondre à la complexité des logiciels modernes. L'informatique des systèmes autonomes (Autonomic computing) s'inscrit dans cette perspective. Elle se propose entre autres de réduire le coût de maintenance des systèmes informatiques en développant des logiciels dits autonomes, c'est à dire dotés de la capacité de s'auto-gérer moyennant une intervention limité d'un opérateur humain. Toutefois, le développement de logiciels autonomes soulèvent de nombreux défis scientifiques et technologiques. Par exemple, l'absence de visibilité de la couche de contrôle dans les applications autonomes rend difficile leur maintenabilité, l'absence d'outils de vérification pour les architectures autonomes est un frein pour l'implémentation d'applications fiables, enfin, la gestion transparente des propriétés non-fonctionnelles et la traçabilité entre le design et l'implémentation sont autant de défis que pose la construction de logiciels autonomes flexibles. La principale contribution de cette thèse est CORONA. CORONA est un canevas logiciel qui vise à faciliter le développement de logiciels autonomes flexibles. Dans cet objectif, CORONA s'appuie sur un langage de description architecturale qui réifie les éléments qui forment la couche de contrôle dans les systèmes autonomes. CORONA permet l'intégration transparente des propriétés non-fonctionnelles dans la description architecturale des systèmes autonomes. il fournit aussi dans sa chaîne de compilation un ensemble d'outils qui permet d'effectuer des vérifications sur l'architecture des systèmes autonomes. Enfin, la traçabilité entre le design et l'implémentation est assurée par un mécanisme de génération des skeletons d'implémentation à partir d'une description architecturale. Les différentes propriétés de CORONA sont illustrées par trois cas d'utilisation.