@phdthesis{Herder1999, author = {Herder, Jens}, title = {A Sound Spatialization Resource Management Framework}, publisher = {University of Tsukuba}, address = {Tsukuba}, organization = {University of Tsukuba}, year = {1999}, abstract = {In a virtual reality environment, users are immersed in a scene with objects which might produce sound. The responsibility of a VR environment is to present these objects, but a practical system has only limited resources, including spatialization channels (mixels), MIDI/audio channels, and processing power. A sound spatialization resource manager, introduced in this thesis, controls sound resources and optimizes fidelity (presence) under given conditions, using a priority scheme based on psychoacoustics. Objects which are spatially close together can be coalesced by a novel clustering algorithm, which considers listener localization errors. Application programmers and VR scene designers are freed from the burden of assigning mixels and predicting sound source locations. The framework includes an abstract interface for sound spatialization backends, an API for the VR environments, and multimedia authoring tools.}, language = {en} } @phdthesis{Deckert2002, author = {Deckert, Carsten}, title = {Wissensorientiertes Projektmanagement in der Produktentwicklung}, publisher = {Shaker Verlag}, address = {Herzogenrath}, year = {2002}, language = {de} } @phdthesis{Neef2002, author = {Neef, Matthias}, title = {Analyse des Schlagfluges durch numerische Str{\"o}mungsberechnung}, volume = {ZLR-Forschungsbericht, 2}, url = {http://nbn-resolving.de/urn:nbn:de:gbv:084-14501}, pages = {245}, year = {2002}, abstract = {Mit Hilfe eines Finite-Volumen-Verfahrens wird die dreidimensionale reibungsbehaftete Stroemung um einen schlagenden Fluegel berechnet. Die Bewegungsparamter sind dabei vergleichbar mit den Bedingungen im Reiseflug grosser Voegel. Das numerische Verfahren wird fuer den gezeigten Anwendungsfall, indem eine Methode zur Realisierung von beliebigen periodischen Bewegungen implemetiert wird. Eine Verifizierung erfolgt durch den Vergleich mit analytischen, numerischen und experimentellen Ergebnissen aus der Literatur. Mit dem verifizierten Verfahren werden die instationaeren Groessen wie Druck und Geschwindigkeiten in der Umgebung des bewegten Fluegels berechnet. Daraus ergeben sich ferner die zeitlich gemittelten Groessen wie Schub und Wirkungsgrad des Schlagfluges. Ausserdem wird die Interaktion von Laengs- und Querwirbeln im Nachlauf des schlagenden Fluegels untersucht. Es zeigt sich ein dominanter Einfluss der pulsierenden Randwirbel aufgrund der endlichen Spannweite, waehrend die Anfahr- und Stoppwirbel, verursacht durch die Fluegelbewegung, im Reiseflug grosser Voegel nahezu vernachlaessigbar erscheinen.}, language = {de} } @phdthesis{Braun2007, author = {Braun, Alexander}, title = {Addressing Single Yb+ Ions: A new scheme for quantum computing in linear ion traps}, publisher = {Cuvillier Verlag}, address = {G{\"o}ttingen}, isbn = {978-3-86727-276-6}, pages = {169}, year = {2007}, language = {en} } @phdthesis{Richter2007, author = {Richter, Nicole}, title = {Interne Kommunikation im Krisenmanagement}, volume = {Personal, Organisation und Arbeitsbeziehungen, 43}, publisher = {Lohmar}, address = {K{\"o}ln}, isbn = {978-3-89936-584-9}, pages = {261}, year = {2007}, language = {de} } @phdthesis{Grote2009, author = {Grote, Wolfgang}, title = {Ein Beitrag zur modellbasierten Regelung von Entnahmedampfturbinen}, publisher = {Bochum}, organization = {Ruhr-Universit{\"a}t Bochum}, pages = {244}, year = {2009}, subject = {Dynamische Modellierung}, language = {de} } @phdthesis{Franz2010, author = {Franz, Thomas}, title = {Semantic Personal Information Management}, publisher = {Universit{\"a}t Koblenz-Landau}, address = {M{\"u}nchen}, isbn = {978-3-86853-706-2}, year = {2010}, language = {en} } @phdthesis{Stuhlsatz2010, author = {Stuhlsatz, Andr{\´e}}, title = {Machine learning with Lipschitz classifiers}, publisher = {Otto-von-Guericke-Universit{\"a}t Magdeburg}, address = {Magdeburg}, doi = {10.25673/5050}, url = {http://nbn-resolving.de/urn:nbn:de:101:1-201104192945}, year = {2010}, abstract = {complex image, like for example that of a known person, and to distinguish it from other objects within half a second. While for a solution of this task the brain has access to a massive parallelism and a vast, hierarchically organized, and auto-associative memory, common computer architectures are just able to a sequential processing of information stored in a non auto-associative memory. Even modern, parallelly operating, multi-processor systems are far away from the performance of our brain. However, nowadays, it is possible to solve complex and memory extensive pattern recognition problems, like the recognition of handwritten digits or the transcription of speech, satisfactorily with a common computer by the use of modern statistical and algorithmic learning approaches. One of the most successful pattern recognition methods is the so-called Support Vector Machine (SVM). The SVM is based on the learning paradigm of structural risk minimization, which outperforms empirical approaches if only few data is available for solving the considered classification problem. Although the SVM has proven very good recognition performances in many cases, the SVM also comes up with limitations, for example if specific a priori knowledge shall be used. In particular, the increasing complexity of applications requires a high adaptivity of the classification method to the specific problem. Also concerning this point, the SVM is limited due to a restricted variety of implementable classification functions. The objective of the present thesis is the development of new learning algorithms for the classification of patterns, that on the one hand overcome the limitations of the SVM, but on the other hand are based on the same theoretical concepts facilitating the good performance of the SVM. Two new algorithms will be presented that are justified by a theoretical generalization of the SVM, and which will be utilized for the first time for a practical implementation. In contrast to the SVM, the new methods make accessible a much larger function class for constructing a classifier. This is an important prerequisite for flexible adaptation of the classifier to difficult classification tasks with particular requirements as well as for the integration of a priori knowledge about the problem at hand. In this work, the way to implementable algorithms leads across different mathematical reformulations of the original problem. Starting with the theoretical generalization of the SVM, it results a restricted optimization problem that is difficult to solve in general. In a first step, this problem is expressed in terms of a restricted minimax-problem by a modification of the suitable classification functions to a still very large function class consisting of (affine-)linear combinations of at least one-time continuously differentiable functions. In the next step, the minimax-problem is converted into a so-called Semi-Infinite Problem (SIP). It turns out, that this particular mathematical problem is appropriate in order to obtain a solution of the original problem for the considered function class using well-known optimization methods. To further exploit the problem structure, an equivalent dual problem is derived from the SIP. Therefore, we prove a duality theorem about the equality of the optimal values of the dual and the original problem. For solving the dual problem, a multilevel iterative approach is developed from which the proposed algorithms follow by pursuing different solution strategies. Moreover, all sub-optimization methods of any stage necessary for an implementation in software are developed. Namely, these are an adapted interior-point-method, a simulated annealing based search heuristics and a particular gradient decent approach. Furthermore, options are depicted for an improvement of efficiency for future implementations. Besides the emphasis on the theoretical development of new learning methods and their practical implementations, all algorithms were implemented in the MATLAB(R) programming environment for the experimental part of the present thesis. Hence, they are also available for further research purposes in future. For the first time, classification results are explored and evaluated in comparison to the SVM on different data sets. As test data, an artificial 2d-dataset as well as two real-world datasets were used. In the concluding experiment, a scenario is prototypically considered to which the SVM is only inadequately applicable and which shall precisely prove the capability of the new methods in that case. It follows, regarding the considered datasets, the proposed learning methods reach comparably good classification accuracy like the SVM in standard applications. Moreover, the particular benefit of the new methods is reflected theoretically and experimentally in the ability to solve classification problems using decision functions that are not accessible to SVMs. Thereby, the underlying ideas, which make the SVM excel compared to other approaches with respect to generalization performances in case of few available learning information, are adequately transported into the proposed new environment. This opens the way for a design and a use of new classifiers that have not been implementable in a robust and generalizing basic concept so far.}, language = {en} } @phdthesis{Steffens2013, author = {Steffens, Jochen}, title = {"Wie viel Realit{\"a}t braucht der Mensch?" - Untersuchungen zum Einfluss der Versuchsumgebung auf die Ger{\"a}uschbeurteilung von Haushaltsger{\"a}ten}, publisher = {Technische Universit{\"a}t}, address = {Berlin}, organization = {Technische Universit{\"a}t}, doi = {10.14279/depositonce-3732}, year = {2013}, language = {de} } @phdthesis{Eilert2013, author = {Eilert, Eva}, title = {New applications of heterologous gene expression in Hansenula polymorpha for protein and metabolite production}, publisher = {Institut f{\"u}r Molekulare Mykologie, Institut f{\"u}r Mikrobiologie}, address = {D{\"u}sseldorf}, organization = {Institut f{\"u}r Molekulare Mykologie, Institut f{\"u}r Mikrobiologie}, url = {http://nbn-resolving.de/urn:nbn:de:hbz:061-20130527-092604-7}, year = {2013}, abstract = {The methylotrophic yeast Hansenula polymorpha has been established as a highly efficient expression system. Strong promoters derived from the methanol metabolism pathway and mitotically stable high-copy integration have made this organism very effective for heterologous protein expression. To improve the processing of secretory proteins targeted by the MFα prepro sequence, variations in the amino acid composition near the processing site of Kex2p were investigated (Manuscript I). The insertion of Ala-Pro between MFα prepro sequence and protein of interest leads to an improved processing of the signal sequence in case of interleukin-6 (IL-6) and interferon-α 2a (IFNα-2a). Modeling of the transition sequences confirmed a better exposition of the Kex2p-cleavage site in dependency of Ala-Pro insertions. Analysis of the secreted proteins revealed activity of an Ala-Pro specific endoprotease. Partial removal of the N-terminal Ala-Pro dipeptide was also detected during IFNα-2a processing. The putative protease responsible for the processing of Ala-Pro dipeptides was identified by gene deletion. Significant similarities with Saccharomyces cerevisiae Ste13p were identified. To generally improve productivity in the H. polymorpha system new fusion protein strategies were investigated (Manuscript II). The recombinantly strongly expressed cytosolic ferritin heavy chain (FTH1) was analyzed with regard to its potential as fusion partner for difficult-to-express target proteins. A fragment of the parathyroid hormone (PTH) was chosen to explore whether fusion to FTH1 enhances PTH stability. We could show that the expression level of this fusion protein was at least of the magnitude of that of FTH1 without fusion partner. The solubility of the fusion protein was increased in comparison to FTH1 alone and no degradation of PTH occurred when fused C-terminally to FTH1. Ectoine and 5-hydroxyectoine belong to the family of compatible solutes which are known to contribute to the adaptation of the cell to osmotic stress by mediation of a constant turgor. H. polymorpha was engineered to express the enzymes of the H. elongata 5-hydroxyectoine biosynthesis pathway (Manuscript III). Expression of the enzymes EctB, EctA, EctC and EctD in H. polymorpha resulted in production of 5-hydroxyectoine in gram per liter scale with almost full conversion of ectoine to 5-hydroxyectoine. We showed that 5-hydroxyectoine synthesis was influenced by feeding of methanol during cultivation. Furthermore, 5-hydroxyectoine synthesis was downregulated by the addition of methionine, lysine or threonine indicating the existence of a feedback regulated aspartate kinase in H. polymorpha.}, language = {en} }