Encoding hieroglyphic texts is time-consuming. If a text already exists as hand-written transcription, there is an alternative, namely OCR. Off-the-shelf OCR systems seem difficult to adapt to the peculiarities of Ancient Egyptian. Presented is a proof-of-concept tool that was designed to digitize texts of Urkunden IV in the hand-writing of Kurt Sethe. It automatically recognizes signs and produces a normalized encoding, suitable for storage in a database, or for printing on a screen or on paper, requiring little manual correction. The encoding of hieroglyphic text is RES (Revised Encoding Scheme) rather than (common dialects of) MdC (Manuel de Codage). Earlier papers argued against MdC and in favour of RES for corpus development. Arguments in favour of RES include longevity of the encoding, as its semantics are font-independent. The present study provides evidence that RES is also much preferable to MdC in the context of OCR. With a well-understood parsing technique, relative positioning of scanned signs can be straightforwardly mapped to suitable primitives of the encoding.

The crystal structures of the LnAgAs2 and LnAuAs2 compounds were reinvestigated by single-crystal diffraction experiments. Contrary to the respective copper compounds, no stuffed variant of the HfCuSi2 type was found. For CeAuAs2, GdAuAs2 and TbAuAs2, a slight under-occupation of the gold position was determined, the other compounds crystallize in a 1:1:2 ratio. Additionally, LaCuAs2 was synthesized for the first time in a 1:1:2 ratio. Due to the fact that imaging plate diffraction systems were used instead of four-circle diffractometers, satellite reflections could be observed for most of the LnCuAs2 compounds (Ln = Ce, Nd, Sm, Gd, Tb, Ho), CeAuAs2, GdAuAs2 and TbAuAs2. Structure models of GdCuAs2, CeAuAs2, GdAuAs2 and TbAuAs2 were developed, rod and layer groups of the respective structural motives were determined and approximants were presented. / Die Kristallstrukturen der LnAgAs2 und LnAuAs2 Verbindungen wurden mittels Einkristallbeugungsexperimenten neu untersucht. Im Gegensatz zu den jeweiligen Kupferverbindungen wurde keine gefüllte Variante des HfCuSi2-Typs gefunden. Im Falle von CeAuAs2, GdAuAs2 und TbAuAs2 wurde eine geringfügige Unterbesetzung der Gold-Position bestimmt, die anderen Verbindungen kristallisieren in einem 1:1:2-Verhältnis. Weiterhin wurde erstmals LaCuAs2 in diesem Verhältnis synthetisiert. Aufgrund der Verwendung von Bildplattensystemen an Stelle von Vierkreisdiffraktometern konnten Satellitenreflexe für den Großteil der LnCuAs2-Verbindungen (Ln = Ce, Nd, Sm, Gd, Tb, Ho), CeAuAs2, GdAuAs2 and TbAuAs2 beobachtet werden. Strukturmodelle von GdCuAs2, CeAuAs2, GdAuAs2 und TbAuAs2 wurden entwickelt, Stab- und Ebenengruppen der einzelnen Strukturmotive wurden bestimmt und Approximanten präsentiert.

Thermische Gehäuseverkrümmungen spielen bei der Dimensionierung von radialen Spielen in Industriedampfturbinen eine große Rolle. Die zuverlässige Vorausberechnung der Gehäuseverkrümmung ist wichtig, um einerseits möglichst kleine Spiele und damit hohe innere Wirkungsgrade erzielen zu können und andererseits Schäden in Folge von Spielüberbrückungen auszuschließen. In der vorliegenden Arbeit wurde die Gehäuseverkrümmung mit analytischen Methoden und mit Hilfe von 3D-FE-Analysen untersucht. Eine neue semi-analytische Berechnungsmethode auf Basis eines Stufenkörpermodells wurde entwickelt. Damit lassen sich komplexe Geometrien und unterschiedliche thermische Randbedingungen berücksichtigen. Der Einfluss von Rotationsasymmetrien auf die Verkrümmung wurde mit Hilfe einer FE-Parameterstudie untersucht und in Form von Einflussfaktoren in das Modell integriert. Im Vergleich zu vereinfachten FE-Modellen und zu gemessenen Verkrümmungen an realen Turbinengehäusen zeigt das semi-analytische Modell eine gute Übereinstimmung. Weiterhin wurde das Verkrümmungsverhalten von Industriedampfturbinengehäusen mit Hilfe von 3D-FE-Analysen untersucht. Durch Abgleich von gemessenen und simulierten Temperaturen und Verformungen konnten Erkenntnisse zum Modellaufbau gesammelt werden. Da die thermischen Randbedingungen bei der Simulation von Gehäuseverkrümmungen von größter Bedeutung sind und Defizite beim Wissensstand zum Wärmeübergang erkannt wurden, sind umfangreiche Messungen des Wärmeübergangskoeffizienten durchgeführt worden. Damit wurde ein Beitrag geleistet, FE-Analysen zukünftig auch bei Dimensionierungs- und Auslegungsfragen einsetzen zu können.

Heat treatment of different steel products have existed for thousands of years. It has always been an important tool to get the microstructure and resulting properties such as hardness and case hardness and it is even more important today than ever before. This project concentrated on the quenching process and means to decrease the distortion caused by this process. The effect of different oils, temperatures, agitation and if gas quenching could give better results were investigated. The results showed that Miller´s 75 quench oil was better than Park´s 420 at slow agitation and that the viscosity of the oils influenced how much changes in agitation speed and oil temperature affected the distortion. It also shows that gas quenching is an alternative to oil quenching if the microstructure can be improved. Otherwise using Miller´s 75 with low agitation in the Surface combustion furnace will give best results.

The present thesis offers a new perspective on Holocaust memory in Greece by examining the ways in which divergent mnemonic representations about the Greek 1940s, as evidenced in polarized public contestation, influence the position of Holocaust in contemporary Greek collective memory. Adopting a micro-level case-study approach, the thesis focuses on the process of renaming a street in Salonika (or Thessaloniki), by examining public discourses around the issue. On the basis of theoretical elaborations in the area of collective memory, and through an application of Kubik and Bernhard's conceptualization of the politics of memory, a qualitative evaluation of Holocaust memory in Salonika is presented by attempting to categorize the memory regime emerging. It is assessed that the memory regime pertaining to the Holocaust is affected by the salience of pre-established memory regimes, occupies a secondary status in the wider mnemonic field and, what is more, is not unified. In such context, a problematic tendency to actually distort the historical record of the Holocaust, in the form of downplaying the complicity of local elites in the implementation of the Nazi genocidal policy, is also detected and explained as a repercussion of the specific dynamics at play whenever political actors engage in discussions about the Greek 1940s. All things considered, the study demonstrates that the official institutionalization of Holocaust memory on a commemorative level, a phenomenon observed during the past twenty years, should not be equated to the emergence of a cosmopolitan Holocaust memory in the country.

Magister Legum - LLM / It has been held that agricultural domestic support would not be such a contentious issue if its only effect was the benefit of local farmers, but this is not the case.1 It was found that several forms of domestic support have the effect of distorting the patterns of agricultural production and trade at an international level, leaving non-supported farmers elsewhere worse off.2 It was thus concluded that such support measures may indeed nullify the benefits which accrue from trade liberalisation and explains how the AoA3 regulates these measures in a way that reduces their trade distorting effects.4 It has been noted that the agricultural sector only accounted for a small percentage of the developed world's Gross Domestic Product {GDP}, yet the regulation of international agricultural trade was not an easy task.5 Smith explains that numerous attempts were made to implement some form of regulation, including a half-hearted effort in the General Agreement on Tariffs and Trade (GATT) and the subsequent AoA upon the creation of the WTO in 1995.6 According to Smith, the successful regulation of international agricultural trade remained elusive, despite Desta MG and McMahon JA explain that the WTO is not very concerned with countries that provide domestic support to their agricultural sectors, as this only matters to the extent that it hopes for liberalising trade in the sector.7 affects trade in that sector.8 It is further observed that the AoA balances out the freedom to provide domestic support with the need to reduce or eliminate the trade distortive effects thereof and note that the AoA has essentially made all forms of domestic support more transparent and easier to deal with.9 A party is therefore unlikely to be challenged, successfully, if domestic support is given in accordance with the provisions of the AoA.10 The aforementioned views only seem to address the merits of the AoA and the way in which it regulates the use of agricultural subsidies. It should however be noted that the literature fails to address the fact that the WTO has not enforced the provisions of the AoA very effectively against the EU and the USA, in light of the continued misuse of subsidies within both parties. In this regard it must be ascertained whether the WTO should impose stricter penalties as a means to deter its member states, especially the EU and USA, from using subsidies in an abusive way. In addition to this, it must be determined which types of penalties can and should be imposed.

The high entropy alloys (HEAs) have been developed based on the concept of entropic stabilization associated with a large number of constituent elements. The high configurational entropy in HEAs is expected to cause promising characteristic properties, i.e. high microstructural stability and high mechanical properties. In this study, the equiatomic fcc-structured FeNiCoCrMn and the bcc-structured TiNbHfTaZr single phase high entropy alloys (HEAs) were investigated regarding the effect of multiple atom species on microstructure, intrinsic lattice strain and mechanical properties. In a comparison with the HEAs, the sub-alloys having less chemical complexity were studied. The selected sub-alloys of the FeNiCoCrMn HEA were FeNiCoCr, FeNiCo, FeNi alloys and pure Ni, while equiatomic TiNbHfTa, TiNbHf, TiNb alloys and pure Nb were studied to compare with the TiNbHfTaZr HEA. The samples in this study were prepared by arc-melting, cold-crucible casting and thermomechanical treatment. The thermal phase stability of the FeNiCoCrMn HEA, TiNbHfTaZr HEA and their sub-alloys were observed and no second phase was formed between 300 - 1500 K. In high entropy alloys, the random arrangement of multiple atom species is assumed to cause large atomic displacements at lattice sites, which give rise to a severe lattice distortion. The evidences of lattice distortion in HEAs have been limitedly reported due to a difficulty of experimental investigation. In this work, the pair distribution function (PDF) method was used to assess local strain with analysis of diffuse intensities on total synchrotron X-ray scattering data. The current study found that the level of local lattice strain associated with atomic displacement was a function of atomic size misfit. The local lattice strain of the FeNiCoCrMn HEA was small and comparable to that of the sub-alloys which obtain similar values of the atomic size misfit. In contrast to the FeNiCoCrMn system, the magnitude of the local lattice strain increased with the value of atomic size misfit from the unary Nb sample to the quinary TiNbHfTaZr HEA. The lattice distortion was evident in the TiNbHfTaZr HEA due to large local lattice strain, but the local lattice strain of the FeNiCoCrMn HEA was not anomalously large. The level of lattice strain determines the solid solution hardening as a consequence of the elastic interaction between dislocations and atoms. The comparable level of the lattice strain in the FeNiCoCrMn HEA, its sub-alloys and Ni sample led to narrow range of hardness (64 – 132 HV) and tensile yield strength (60 – 192 MPa). For the bcc-structured samples, the hardness and the yield strength significantly varied depending on the level of local lattice strain, between 80 – 327 HV of hardness and 207 – 985 MPa of tensile yield strength. It is clear from the result that the atomic size misfit influences the level of the local lattice strain and the solid solution hardening. Cold rotary swaging was used to study the work hardening in the HEAs because it can delay fracture by large hydrostatic stresses. The large plastic deformability was observed in the FeNiCoCrMn and TiNbHfTaZr HEAs. The TiNbHfTaZr HEA was cold-swaged by 90% reduction of the cross-sectional area without intermediate annealing. The FeNiCoCrMn HEA was swaged until 85% reduction of the cross-sectional area; however, it was observed that it could be further deformed. The dislocation densities of the HEAs and its sub-alloys after the cold deformation were calculated as in the range between 1014 - 1015 m-2, in a good agreement with reported values of conventional metals after severe plastic deformation. This finding suggested that the level of dislocation density storage was correlated with the number of the constituent elements, the level of lattice distortion associated with atomic size misfit and the intrinsic properties (i.e. the stacking fault energy and the melting point). Whereas the intrinsic lattice strains of the FeNiCoCrMn HEA and its sub-alloys were comparable, the levels of dislocation storage were different possibly due to a difference of stacking fault energy. For the bcc-structured samples, the dislocation densities of the TiNbHfTaZr HEA, TiNbHfTa and TiNbHf alloys were large due to the large atomic size misfits. The high dislocation density leads to strong interactions between dislocations, which results in high resistance to dislocation motions. The high mechanical properties (hardness and yield strength) in the as-deformed FeNiCoCrMn and TiNbHfTaZr HEA were presented with the evidence of high dislocation densities. Moreover, the hardness and yield strength of the FeNiCoCrMn HEA significantly increased by the deformation, while those of the TiNbHfTaZr HEA after the deformation were slightly changed from the undeformed HEA. The large work hardenability of the FeNiCoCrMn HEA is possibly caused by small solid solution hardening and ease of twin formation. The research results suggest a further step towards designing an alloy composition for a development of microstructure and mechanical properties of high entropy alloys. It is evidently clear from the findings that the large number of constituent elements (in a term of high configurational entropy) is not only a factor in the determination of lattice distortion, microstructure and mechanical properties, but the type and the combination of constituent elements including the atomic interactions (i.e. atomic size misfit) have also an effect.:Abstract v Zusammenfassung ix Contents xiii 1. Motivation and objectives 1 2. Fundamentals 5 2.1 Concept of high entropy alloys 5 2.1.1 Phase formation and thermodynamic 5 2.1.2 Four core effects 10 2.2 Alloy classification of high entropy alloys 13 2.3 Mechanical properties of high entropy alloys 14 3. Experiments 19 3.1 Alloy preparation 19 3.1.1 Alloy selection 19 3.1.2 Melting and casting 21 3.1.3 Thermomechanical treatment 23 3.2 Sample characterization 27 3.2.1 Chemical analyses 27 3.2.2 Differential scanning calorimetry (DSC) 27 3.2.3 Scanning electron microscopy and microstructural analyses 28 3.2.4 X-ray diffraction (XRD) 29 3.2.5 High energy synchrotron X-ray diffraction 29 3.2.6 Mechanical Properties 33 4. Thermal phase stability of single phase high entropy alloys 35 5. An assessment of lattice strain in single phase high entropy alloys 49 5.1 Analysis of micro lattice strain on fcc- and bcc-structured high entropy alloys 50 5.2 Analysis of local lattice strain on fcc- and bcc-structured high entropy alloys 56 6. Solid solution hardening in single phase high entropy alloys 65 6.1 Hardness of fcc- and bcc-structured high entropy alloys 65 6.2 Tensile strength of fcc- and bcc-structured high entropy alloys 70 6.3 Correlation between atomic size misfit and solid solution hardening in Ti-Nb-Hf-Ta-Zr system 82 7. Work hardening in single phase high entropy alloys 91 7.1 Work hardenability of fcc- and bcc-structured high entropy alloys 91 7.2 Dislocation density of fcc- and bcc-structured high entropy alloys after cold swaging 93 8. Summary and outlook 109 8.1 Summary 109 8.2 Outlook 112 References 113 Acknowledgements 131 Erklärung 133 / Die Hochentropielegierungen (HELen) wurden auf der Grundlage des Konzepts der entropischen Stabilisierung entwickelt, was eine große Anzahl von Legierungselementen beinhaltet. Es wird erwartet, dass die hohe Konfigurationsentropie in HELen vielversprechende charakteristische Eigenschaften hervorruft, d.h. hohe mikrostrukturelle Stabilität und hohe mechanische Eigenschaften. In dieser Studie wurden die äquiatomare kfz-strukturierte FeNiCoCrMn und die krz-strukturierte TiNbHfTaZr Einphasen-Hochentropielegierung hinsichtlich der Wirkung mehrerer Atomarten auf das Gefüge, die intrinsische Gitterdehnung und die mechanischen Eigenschaften untersucht. Im Vergleich zu den HELen wurden die Sublegierungen mit geringerer chemischer Komplexität untersucht. Die ausgewählten Sublegierungen der FeNiCoCrMn HEL waren FeNiCoCr, FeNiCo, FeNi-Legierungen und reines Ni, während äquiatomare TiNbHfTa, TiNbHf, TiNbHf, TiNb-Legierungen und reines Nb im Vergleich zur TiNbHfTaZr HEL untersucht wurden. Die Proben in dieser Studie wurden durch Lichtbogenschmelzen, Kalttiegelguss und thermomechanische Behandlung hergestellt. Die thermische Phasenstabilität der FeNiCoCrMn HEL, der TiNbHfTaZr HEL und ihrer Sublegierungen wurde untersucht und es wurde keine zweite Phase zwischen 300 - 1500 K gebildet. Bei Hochentropielegierungen wird angenommen, dass die zufällige Anordnung mehrerer Atomarten zu großen Atomverschiebungen an den Gitterplätzen führt, die eine starke Gitterverzerrung hervorrufen. Aufgrund der Schwierigkeit der experimentellen Untersuchung wurden Beweise für Gitterverzerrungen bei HELen nur begrenzt berichtet. In dieser Studie wurde die Methode der Paarverteilungsfunktion (PDF) verwendet, um die lokale Dehnung mit Analyse der diffusen Intensitäten der gesamten Synchrotron-Röntgenstreuungsdaten zu beurteilen. Die aktuelle Studie ergab, dass die Höhe der lokalen Gitterdehnung, die mit der atomaren Verschiebung einhergeht, eine Funktion der Differenz der Atomgröße ist. Die lokale Gitterdehnung der FeNiCoCrMn HEL war klein und vergleichbar mit der der Sublegierungen, für die ähnliche Werte für die Atomgrößen-Unterschiede ermittelt wurden. Im Gegensatz zum FeNiCoCrMn-System stieg die Größe der lokalen Gitterdehnung mit dem Wert der Atomgrößendifferenz von der unären Nb-Probe bis zur quinären TiNbHfTaZr HEL. Die Gitterverzerrung war in der TiNbHfTaZr HEL aufgrund der großen lokalen Gitterdehnung offensichtlich, wohingegen die lokale Gitterdehnung der FeNiCoCrMn HEL nicht ungewöhnlich groß war. Die Höhe der Gitterdehnung bestimmt die Mischkristallverfestigung als Folge der elastischen Wechselwirkung zwischen Versetzungen und Atomen. Der vergleichbare Wert der Gitterdehnung in der FeNiCoCrMn HEL, seinen Sublegierungen und den Ni-Proben führte zu einem engen Härte- (64 - 132 HV) und Streckgrenzenbereich (60 - 192 MPa). Für die krz-strukturierten Proben variierten die Härte und die Streckgrenze dagegen je nach Höhe der lokalen Gitterdehnung signifikant, d.h zwischen 80 - 327 HV hinsichtlich der Härte und zwischen 207 - 985 MPa hinsichtlich der Streckgrenze. Aus dem Ergebnis ist ersichtlich, dass die Differenz der Atomgröße einen Einfluss auf die Höhe der lokalen Gitterdehnung und die Mischkristallverfestigung hat. Das Kalthämmen wurde für die Untersuchung der Kaltverfestigung in den HELen genutzt, da es den Bruch durch die großen hydrostatischen Spannungen verzögern kann. Die große plastische Verformbarkeit wurde bei den FeNiCoCrMn und TiNbHfTaZr HELen beobachtet. Die TiNbHfTaZr HEL wurde ohne Zwischenglühen um 90% der Querschnittsfläche kaltgehämmert. Die FeNiCoCrMn HEL wurde bis zu einer Verkleinerung der Querschnittsfläche von 85% gehämmert, wobei jedoch eine weitere Verformung möglich gewesen wäre. Die Versetzungsdichten der HELen und ihrer Sublegierungen wurden nach dem Verformung in einem Bereich zwischen 1014 - 1015 m-2 berechnet, was in guter Übereinstimmung mit den berichteten Werten konventioneller Metalle nach starker plastischer Verformung ist. Dieses Ergebnis deutete darauf hin, dass die Höhe der gespeicherten Versetzungsdichte mit der Anzahl der beinhaltenden Elemente, dem Grad der Gitterverzerrung im Zusammenhang mit der Differenz der Atomgröße und den intrinsischen Eigenschaften (d.h. der Stapelfehlerenergie und dem Schmelzpunkt) korreliert. Obwohl die intrinsischen Gitterdehnungen der FeNiCoCrMn HEL und seiner Sublegierungen vergleichbar waren, waren die Werte der gespeicherten Versetzungen unterschiedlich, was möglicherweise an einer Differenz der Stapelfehlerenergie lag. Für die krz-strukturierten Proben waren die Versetzungsdichten der TiNbHfTaZr HEL, der TiNbHfTa- und der TiNbHf-Legierungen aufgrund der großen Atomgrößenunterschiede hoch. Die hohe Versetzungsdichte bewirkt starke Wechselwirkungen zwischen den Versetzungen, was zu einem hohen Widerstand gegen Versetzungsbewegungen führt. Die hohen mechanischen Eigenschaften (Härte und Streckgrenze) in den verformten FeNiCoCrMn und TiNbHfTaZr HELen wurden mit dem Nachweis hoher Versetzungsdichten belegt. Darüber hinaus wurden die Härte und die Streckgrenze des FeNiCoCrMn HEL durch das Kalthämmern deutlich erhöht, während die der TiNbHfTaZr HEL nach dem Hämmerprozess nur leicht gegenüber der unverformten HEL verändert wurden. Die große Kaltverfestigung der FeNiCoCrMn HEL ist möglicherweise auf eine geringe Mischkristallhärtung und eine geringfügige Zwillingsbildung zurückzuführen. Die Forschungsergebnisse empfehlen für die Entwicklung des Gefüges und der mechanischen Eigenschaften von Hochentropielegierungen weitere Schritte hinsichtlich eines zielführenden Legierungsdesigns durchzuführenhin. Aus den Ergebnissen geht eindeutig hervor, dass die große Anzahl an Legierungselementen ( hinsichtlich einer hochkonfigurativen Entropie) nicht die einzige Einflussgrößebei der Bestimmung von Gitterverzerrungen, dem Gefüge und der mechanischen Eigenschaften darstellt, sondern auch die Art und die Kombination der Legierungselementen einschließlich der atomaren Wechselwirkungen (d.h. Atomgrößenunterschiede) einen Effekt haben.:Abstract v Zusammenfassung ix Contents xiii 1. Motivation and objectives 1 2. Fundamentals 5 2.1 Concept of high entropy alloys 5 2.1.1 Phase formation and thermodynamic 5 2.1.2 Four core effects 10 2.2 Alloy classification of high entropy alloys 13 2.3 Mechanical properties of high entropy alloys 14 3. Experiments 19 3.1 Alloy preparation 19 3.1.1 Alloy selection 19 3.1.2 Melting and casting 21 3.1.3 Thermomechanical treatment 23 3.2 Sample characterization 27 3.2.1 Chemical analyses 27 3.2.2 Differential scanning calorimetry (DSC) 27 3.2.3 Scanning electron microscopy and microstructural analyses 28 3.2.4 X-ray diffraction (XRD) 29 3.2.5 High energy synchrotron X-ray diffraction 29 3.2.6 Mechanical Properties 33 4. Thermal phase stability of single phase high entropy alloys 35 5. An assessment of lattice strain in single phase high entropy alloys 49 5.1 Analysis of micro lattice strain on fcc- and bcc-structured high entropy alloys 50 5.2 Analysis of local lattice strain on fcc- and bcc-structured high entropy alloys 56 6. Solid solution hardening in single phase high entropy alloys 65 6.1 Hardness of fcc- and bcc-structured high entropy alloys 65 6.2 Tensile strength of fcc- and bcc-structured high entropy alloys 70 6.3 Correlation between atomic size misfit and solid solution hardening in Ti-Nb-Hf-Ta-Zr system 82 7. Work hardening in single phase high entropy alloys 91 7.1 Work hardenability of fcc- and bcc-structured high entropy alloys 91 7.2 Dislocation density of fcc- and bcc-structured high entropy alloys after cold swaging 93 8. Summary and outlook 109 8.1 Summary 109 8.2 Outlook 112 References 113 Acknowledgements 131 Erklärung 133

<p>In this work, we study a class of Markovian information theoretic optimization problems motivated by the recent interests in incorporating mutual information as performance metrics which gives evident success in representation learning, feature extraction and clustering problems. In particular, we focus on the information bottleneck (IB) and privacy funnel (PF) methods and their recent multi-view, multi-source generalizations that gain attention because the performance significantly improved with multi-view, multi-source data. Nonetheless, the generalized problems challenge existing IB and PF solves in terms of the complexity and their abilities to tackle large-scale data. </p> <p>To address this, we study both the IB and PF under a unified framework and propose solving it through splitting methods, including renowned algorithms such as alternating directional method of multiplier (ADMM), Peaceman-Rachford splitting (PRS) and Douglas-Rachford splitting (DRS) as special cases. Our convergence analysis and the locally linear rate of convergence results give rise to new splitting method based IB and PF solvers that can be easily generalized to multi-view IB, multi-source PF. We implement the proposed methods with gradient descent and empirically evaluate the new solvers in both synthetic and real-world datasets. Our numerical results demonstrate improved performance over the state-of-the-art approach with significant reduction in complexity. Furthermore, we consider the practical scenario where there is distribution mismatch between training and testing data generating processes under a known bounded divergence constraint. In analyzing the generalization error, we develop new techniques inspired by the input-output mutual information approach and tighten the existing generalization error bounds.</p>

The new generation of giant ground-based telescopes will see their first light this decade. These state-of-the-art facilities will significantly surpass the resolving power of modern space-based observatories such as the James Webb telescope, thanks to their enormous aperture size and adaptive optics (AO) facilities. Without AO, atmospheric turbulence would degrade the image quality of these enormous telescopes to that of a 50 cm amateur one. These extremely large telescopes (ELTs) will further benefit from a particular branch of AO called multi-conjugate adaptive optics (MCAO), which provides an extremely high resolving power over a much wider field of view as compared to classical AO systems. The design and fabrication of such systems, as well as their optimal use for science operation, pose a great challenge as they are an order of magnitude more complicated than current AO systems. To face such a challenge, the combined knowledge of MCAO system design and fabrication, working in tandem with scientific insights into new astronomy science cases, is an extremely valuable and essential pairing. This thesis is an effort to not only contribute to the design and fabrication of ELT MCAO facilities, but also provide guidance on the optimal method to utilize these giant telescopes to achieve unprecedented astrometric measurements. On the instrumentation side, in partnership with the National Research Council of Canada's - Herzberg Astronomy and Astrophysics Institute as well as W.M. Keck Observatory in Hawaii, I was involved in the design and fabrication of a cutting edge new wavefront sensor, which is the eye of an AO system. I performed opto-mechanical design and verification studies for components of the Keck infrared pyramid wavefront sensor (IR-PWFS) as well as the Keck Planet Imager and characterizer (KPIC) instrument, which have both been commissioned and are in science operation. Furthermore, I designed the alignment plan and participated in the modification and alignment operation of a few components on the Keck II adaptive optics bench on the summit of Mauna Kea. To pave the way for the design verification of future MCAO systems for ELTs, I proposed a new method for an old challenge in the path of AO system design and verification: a flexible method for precise intensity pattern injection into laboratory AO benches. AO benches are the backbone of instrument design and modeling. One of the challenges especially important for the future generation of MCAO systems for ELTs is the verification of the effect of shadowed regions on the primary mirror. During my PhD, I successfully demonstrated the feasibility of a new proposed method to accurately model the telescope pupil. This work was done in partnership with the Laboratoire d'Astrophysique de Marseille (LAM) in France. The method I developed at LAM will be implemented in the AO Lab at NRC Herzberg Astronomy and Astrophysics. As an observational astronomer, I focused on developing methods for making optimal astrometric measurements with MCAO-enabled telescopes. The expected unparalleled astrometric precision of ELTs comes with many unprecedented challenges that if left unresolved, would jeopardize the success of these facilities as they would not be able to reach their science goals. I used observations with the only available MCAO system in science operation, the Gemini MCAO system on the 8-meter Gemini South telescope in Chile, to develop and verify a pipeline specifically designed for very high-precision astrometric studies with MCAO-fed imagers. I successfully used the pipeline to provide the precise on-sky differential distortion of the Gemini South telescope and its MCAO facilities by looking deep into the core of globular cluster NGC~6723. Using this pipeline, I produced high quality proper motions with an uncertainty floor of $\sim 45$\,$\mu$as~yr$^{-1}$ as well as measured the proper motion dispersion profile of NGC~6723 from a radius of $\sim 10$ arcseconds out to $\sim 1$\,arcminute, based on $\sim 12000$ stars. I also produced a high-quality optical-near-infrared color magnitude diagram which clearly shows the extreme horizontal branch and main-sequence knee of this cluster. / Graduate

The increasing propagation of power electronics technologies and the related operational equipment such as inverter-fed drives and solid-state transformers in electrical grids lead to growing harmonic distortion of the voltage waveform. Harmonics modify the shape of a voltage waveform but also affect the peak voltage despite constant RMS voltage. Resulting from this, it was previously found that harmonics not only have an influence on the partial discharge patterns and characteristics but also the lifetime of model insulations. The epoxy-mica main wall insulation of rotating electrical machines is partial discharge-resistant to certain extent. Nonetheless, severe production faults should be detected by partial discharge measurements as they may lead to accelerated ageing. The purpose of this contribution is to investigate the impact of harmonics on the partial discharge characteristics of the insulation system of rotating electrical machines in general and especially on a test object with an artificial void fault. An assessment of the partial discharge pattern and apparent charge measurement is given by analyzing the characteristics to harmonic waveform parameters that incoporate peak alteration and shape modification. It is shown that the voltage shape and gradients play a subordinate role compared to the voltage peak altering effect of harmonics. The effects are more pronounced with the test object with an artificially introduced fault.