In this paper we propose an approach for incremental learning of semi-supervised SVM. The proposed approach makes use of the locality of radial basis function kernels to do local and incremental training of semi-supervised support vector machines. The algorithm introduces a se- quential minimal optimization based implementation of the branch and bound technique for training semi-supervised SVM problems. The novelty of our approach lies in the in the introduction of incremental learning techniques to semisupervised SVMs.

In this master thesis multiple time series of shipping industry and financial data are analysed in order to create a forecasting model to forecast freight rate indices. The data of main interest which are predicted are the two freight rate indices, BDI and BDTI, from the Baltic Exchange. The project investigates the possibilities for aggregated Vector Autoregression(VAR) models to outperform simple univariate models, in this case, an Autoregressive Integrated Moving Average(ARIMA) with seasonal components. The other part of this thesis is to model market shocks in the freight rate indices, given impulses in the other underlying VAR-model time series using the impulse response function. The main results are that the VAR-model forecast outperforms the ARIMA-model in forecasting the tanker freight rate index (BDTI), while the the bulk freight rate index(BDI) is better predicted by the simple ARIMA when calculating the forecast mean square error. / I denna avhandling analyseras multipla tidsserier över rederinärings- och finansiell data i syfte att skapa en prognosticerande modell för att prognosticera fraktratsindex. Dataserierna som i huvudsak prognosticeras är fraktratsindexen BDI och BDTI från Baltic exchange. I projektet undersöks om en aggregerad Vektor Autoregressiv(VAR) modell överträffar en univariat modell, i detta fall en Autoregressive Integrated Moving Average(ARIMA) med säsongsvariabel. I andra delen av denna avhandling modelleras chocker i fraktratsindexen givet impulser i de andra underliggande tidsserierna i de aggregerade VAR-modellerna. Huvudresultaten är att VAR-modellens prognos överträffar ARIMA-modellen för tankerraterna (BDTI), medan bulkraterna(BDI) bättre prognosticeras av ARIMA-modellen, i avseende på prognosernas beräknade mean square error.

To answer the question who spoke when, speaker diarization (SD) is a critical step for many speech applications in practice. The task of our project is building a MFCC-vector based speaker diarization system on top of a speaker verification system (SV), which is an existing Call-centers application to check the customer’s identity from a phone call. Our speaker diarization system uses 13-Dimensional MFCCs as Features, performs Voice Active Detection (VAD), segmentation, Linear Clustering and the Hierarchical Clustering based on GMM and the BIC score. By applying it, we decrease the Equal Error Rate (EER) of the SV from 18.1% in the baseline experiment to 3.26% on the general call-center conversations. To better analyze and evaluate the system, we also simulated a set of call-center data based on the public audio databases ICSI corpus. / För att svara på frågan vem som talade när är högtalardarisering (SD) ett kritiskt steg för många talapplikationer i praktiken. Uppdraget med vårt projekt är att bygga ett MFCC-vektorbaserat högtalar-diariseringssystem ovanpå ett högtalarverifieringssystem (SV), som är ett befintligt Call-center-program för att kontrollera kundens identitet från ett telefonsamtal. Vårt högtalarsystem använder 13-dimensionella MFCC: er som funktioner, utför Voice Active Detection (VAD), segmentering, linjär gruppering och hierarkisk gruppering baserat på GMM och BIC-poäng. Genom att tillämpa den minskar vi EER (Equal Error Rate) från 18,1 % i baslinjeexperimentet till 3,26 % för de allmänna samtalscentret. För att bättre analysera och utvärdera systemet simulerade vi också en uppsättning callcenter-data baserat på de offentliga ljuddatabaserna ICSI corpus.

Foamy viruses (FVs) or heterologous retroviruses pseudotyped with FV glycoprotein enable transduction of a great variety of target tissues of disparate species. Specific cellular entry receptors responsible for this exceptionally broad tropism await their identification. Though, ubiquitously expressed heparan sulfate proteoglycan (HS-PG) is known to serve as an attachment factor of FV envelope (Env)-containing virus particles, greatly enhancing target cell permissiveness. Production of high-titer, FV Env-containing retroviral vectors is strongly dependent on the use of cationic polymer-based transfection reagents like polyethyleneimine (PEI). We identified packaging cell-surface HS-PG expression to be responsible for this requirement. Efficient release of FV Env-containing virus particles necessitates neutralization of HS-PG binding sites by PEI. Remarkably, remnants of PEI in FV Env-containing vector supernatants, which are not easily removable, negatively impact target cell transduction, in particular those of myeloid and lymphoid origin. To overcome this limitation for production of FV Env-containing retrovirus supernatants, we generated 293T-based packaging cell lines devoid of HS-PG by genome engineering. This enabled, for the first, time production of inhibitor-free, high-titer FV Env-containing virus supernatants by non-cationic polymer-mediated transfection. Depending on the type of virus, produced titers were 2- to 10-fold higher compared with those obtained by PEI transfection.

In computer graphics we can handle unstructured triangular 3D meshes which are not too usable for processing through their irregularity. In these situations it occurs need of conversion that 3D mesh to more suitable representation. Some kind of 3D spline surface can be proper alternative because it institutes regularity in the form of control points grid and that's why it is more suitable for next processing. During conversion, which is described in this thesis, quadrilateral 3D mesh is constructed at first. This mesh has regular structure but mainly the structure corresponds to structure of control points grid of resulting 3D spline surface. Created quadrilateral 3D mesh can be saved and consequently used in specific modeling applications for T-spline surface creation.

The aim of this thesis is to present new results concerning duality in scalar optimization. We show how the theory can be applied to optimization problems arising in the theory of risk measures, portfolio optimization and machine learning. First we give some notations and preliminaries we need within the thesis. After that we recall how the well-known Lagrange dual problem can be derived by using the general perturbation theory and give some generalized interior point regularity conditions used in the literature. Using these facts we consider some special scalar optimization problems having a composed objective function and geometric (and cone) constraints. We derive their duals, give strong duality results and optimality condition using some regularity conditions. Thus we complete and/or extend some results in the literature especially by using the mentioned regularity conditions, which are weaker than the classical ones. We further consider a scalar optimization problem having single chance constraints and a convex objective function. We also derive its dual, give a strong duality result and further consider a special case of this problem. Thus we show how the conjugate duality theory can be used for stochastic programming problems and extend some results given in the literature. In the third chapter of this thesis we consider convex risk and deviation measures. We present some more general measures than the ones given in the literature and derive formulas for their conjugate functions. Using these we calculate some dual representation formulas for the risk and deviation measures and correct some formulas in the literature. Finally we proof some subdifferential formulas for measures and risk functions by using the facts above. The generalized deviation measures we introduced in the previous chapter can be used to formulate some portfolio optimization problems we consider in the fourth chapter. Their duals, strong duality results and optimality conditions are derived by using the general theory and the conjugate functions, respectively, given in the second and third chapter. Analogous calculations are done for a portfolio optimization problem having single chance constraints using the general theory given in the second chapter. Thus we give an application of the duality theory in the well-developed field of portfolio optimization. We close this thesis by considering a general Support Vector Machines problem and derive its dual using the conjugate duality theory. We give a strong duality result and necessary as well as sufficient optimality conditions. By considering different cost functions we get problems for Support Vector Regression and Support Vector Classification. We extend the results given in the literature by dropping the assumption of invertibility of the kernel matrix. We use a cost function that generalizes the well-known Vapnik's ε-insensitive loss and consider the optimization problems that arise by using this. We show how the general theory can be applied for a real data set, especially we predict the concrete compressive strength by using a special Support Vector Regression problem.

The aim of this thesis is to present new results concerning duality in scalar optimization. We show how the theory can be applied to optimization problems arising in the theory of risk measures, portfolio optimization and machine learning. First we give some notations and preliminaries we need within the thesis. After that we recall how the well-known Lagrange dual problem can be derived by using the general perturbation theory and give some generalized interior point regularity conditions used in the literature. Using these facts we consider some special scalar optimization problems having a composed objective function and geometric (and cone) constraints. We derive their duals, give strong duality results and optimality condition using some regularity conditions. Thus we complete and/or extend some results in the literature especially by using the mentioned regularity conditions, which are weaker than the classical ones. We further consider a scalar optimization problem having single chance constraints and a convex objective function. We also derive its dual, give a strong duality result and further consider a special case of this problem. Thus we show how the conjugate duality theory can be used for stochastic programming problems and extend some results given in the literature. In the third chapter of this thesis we consider convex risk and deviation measures. We present some more general measures than the ones given in the literature and derive formulas for their conjugate functions. Using these we calculate some dual representation formulas for the risk and deviation measures and correct some formulas in the literature. Finally we proof some subdifferential formulas for measures and risk functions by using the facts above. The generalized deviation measures we introduced in the previous chapter can be used to formulate some portfolio optimization problems we consider in the fourth chapter. Their duals, strong duality results and optimality conditions are derived by using the general theory and the conjugate functions, respectively, given in the second and third chapter. Analogous calculations are done for a portfolio optimization problem having single chance constraints using the general theory given in the second chapter. Thus we give an application of the duality theory in the well-developed field of portfolio optimization. We close this thesis by considering a general Support Vector Machines problem and derive its dual using the conjugate duality theory. We give a strong duality result and necessary as well as sufficient optimality conditions. By considering different cost functions we get problems for Support Vector Regression and Support Vector Classification. We extend the results given in the literature by dropping the assumption of invertibility of the kernel matrix. We use a cost function that generalizes the well-known Vapnik's ε-insensitive loss and consider the optimization problems that arise by using this. We show how the general theory can be applied for a real data set, especially we predict the concrete compressive strength by using a special Support Vector Regression problem.

Die vorliegende Arbeit beschaeftigt sich mit dem strukturellen Aufbau von (1-100) M-plane GaN, das mit plasmaunterstuetzter Molekularstrahlepitaxie auf gamma-LiAlO2(100) Substraten gewachsen wurde. Die heteroepitaktische Ausrichtung einerseits, sowie die Mikrostruktur und die Erzeugungsmechanismen der Defekte andererseits, wurde mit der Transmissionselektronenemikroskopie (TEM) systematisch untersucht. Das gamma-LiAlO2 Substrat reagiert heftig im Mikroskop unter Bestrahlung mit hochenergetischen Elektronen. Waehrend dieser Strahlenschaedigung verliert das Material seine urspruengliche kristalline Struktur und vollzieht eine Phasentransformation, die anhand einer Serie von Feinbereichsbeugungsdiagrammen nachgewiesen werden konnte. Die atomare Grenzflaechenstruktur zwischen epitaktisch gewachsenem alpha-GaN(1-100) und tetragonalem gamma-LiAlO2 Substrat ist mittels HRTEM untersucht worden. Die neuartige Epitaxiebeziehung ist mit Elektronenbeugung bestaetigt worden und lautet folgendermassen: (1-100)GaN liegt parallel zu (100)gamma-LiAlO2 und [11-20]GaN ist parallel zu [001]gamma-LiAlO2. Die Realstruktur der M-plane GaN Schichten, die auf (100)gamma-LiAlO2 gewachsen werden, unterscheidet sich erheblich von der in C-plane Orientierung hergestellten Epischichten. Ausfuehrliche TEM Untersuchungen zeigen, dass die M-plane Schichten vor allem intrinsische (I1 und I2) und extrinsische (E) Stapelfehler in der Basalebene enthalten. Der vorherrschende I2 Stapelfehler besitzt keine Komponente des Verschiebungsvektors senkrecht zur Ebene und ist damit nicht geeignet, epitaktische Dehnung entlang der [11-20] Richtung abzubauen. Darueberhinaus ist eine komplexe Grenze in der (10-10) Prismen- flaeche entdeckt worden, die zur Grenzflaeche geneigt verlaeuft. Die Defekte in den M-plane GaN Epischichten werden waehrend der anfaenglichen Keimbildungsphase erzeugt. Atomare Stufen entlang der [001] Richtung auf dem LiAlO2 Substrat fuehren zur Bildung von Stapelfehlern vom Typ I2. / In this work the structure of (1-100)M-plane GaN epitaxially grown on gamma-LiAlO2(100) by using plasmaassisted molecular beam epitaxy (PAMBE) is studied. The heteroepitaxial alignment and the microstructure of M-plane GaN as well as the defect formation in the layer are systematically investigated by using transmission electron microscopy (TEM). The gamma-LiALO2 substrate reacts under irradiation of high-energy electrons in the TEM (200-300 keV).The material looses its original crystalline structure during this process undergoing irradiation damage followed by a phase transformation as it is verified by a series of selected area diffraction patterns taken under constant electron dose. The result is a structural phase transformation from the tetragonal gamma to the trigonal alpha phase. The atomic interface structure of epitaxially grown hexagonal alpha-GaN(1-100) layers on tetragonal gamma-LiAlO2 (100) substrates is investigated by means of HRTEM. The novel epitaxial orientation relationship verified by electron diffraction is given by (1-100)GaN parallel to (100)gamma-LiAlO2 and [11-20]GaN parallel to [001]gamma-LiAlO2. The defect structure of M-plane GaN epilayers grown on gamma-LiAlO2(100) substrates is different to that of C-plane GaN. Our detailed TEM studies reveal that the M-plane layers mainly contain intrinsic I1 and I2 and extrinsic E basal plane stacking faults. The dominant I2 stacking fault has no out-of-plane displacement vector component and is thus not qualified for epitaxial strain relief along the [11-20] axis. Beyond this, a complex type of planar defect is detected in the (10-10) prism plane which is inclined with respect to the interface. The study of nucleation samples shows that the surface morphology is directly correlated to the generation of the dominant planar defects. Atomic steps along the [001] direction in the gamma-LiAlO2 substrate result in the formation of basal plane stacking faults I2.

In dieser Arbeit werden zwei wesentliche neue Ergebnisse vorgestellt. Das erste bezieht sich auf die olfaktorische Kodierung und das zweite auf das sensorische Gedaechtnis. Beide Phaenomene werden am Beispiel des Gehirns der Honigbiene untersucht. In Bezug auf die olfaktorische Kodierung zeige ich, dass die neuronale Dynamik waehrend der Stimulation im Antennallobus duftspezifische Trajektorien beschreibt, die in duftspezifischen Attraktoren enden. Das Zeitinterval, in dem diese Attraktoren erreicht werden, betraegt unabhaengig von der Identitaet und der Konzentration des Duftes ungefaehr 800 ms. Darueber hinaus zeige ich, dass Support-Vektor Maschinen, und insbesondere Perzeptronen, ein realistisches und biologisches Model der Wechselwirkung zwischen dem Antennallobus (dem kodierenden Netwerk) und dem Pilzkoerper (dem dekodierenden Netzwerk) darstellen. Dieses Model kann sowohl Reaktionszeiten von ca. 300 ms als auch die Invarianz der Duftwahrnehmung gegenueber der Duftkonzentration erklaeren. In Bezug auf das sensorische Gedaechtnis zeige ich, dass eine einzige Stimulation ohne Belohnung dem Hebbschen Postulat folgend Veraenderungen der paarweisen Korrelationen zwischen Glomeruli induziert. Ich zeige, dass diese Veranderungen der Korrelationen bei 2/3 der Bienen ausreichen, um den letzten Stimulus zu bestimmen. In der zweiten Minute nach der Stimulation ist eine erfolgreiche Bestimmung des Stimulus nur bei 1/3 der Bienen moeglich. Eine Hauptkomponentenanalyse der spontanen Aktivitaet laesst erkennen, dass das dominante Muster des Netzwerks waehrend der spontanen Aktivitaet nach, aber nicht vor der Stimulation das duftinduzierte Aktivitaetsmuster bei 2/3 der Bienen nachbildet. Man kann deshalb die duftinduzierten (Veraenderungen der) Korrelationen als Spuren eines Kurzzeitgedaechtnisses bzw. als Hebbsche "Reverberationen" betrachtet werden. / Two major novel results are reported in this work. The first concerns olfactory coding and the second concerns sensory memory. Both phenomena are investigated in the brain of the honeybee as a model system. Considering olfactory coding I demonstrate that the neural dynamics in the antennal lobe describe odor-specific trajectories during stimulation that converge to odor-specific attractors. The time interval to reach these attractors is, regardless of odor identity and concentration, approximately 800 ms. I show that support-vector machines and, in particular perceptrons provide a realistic and biological model of the interaction between the antennal lobe (coding network) and the mushroom body (decoding network). This model can also account for reaction-times of about 300 ms and for concentration invariance of odor perception. Regarding sensory memory I show that a single stimulation without reward induces changes of pairwise correlation between glomeruli in a Hebbian-like manner. I demonstrate that those changes of correlation suffice to retrieve the last stimulus presented in 2/3 of the bees studied. Succesful retrieval decays to 1/3 of the bees within the second minute after stimulation. In addition, a principal-component analysis of the spontaneous activity reveals that the dominant pattern of the network during the spontaneous activity after, but not before stimulation, reproduces the odor-induced activity pattern in 2/3 of the bees studied. One can therefore consider the odor-induced (changes of) correlation as traces of a short-term memory or as Hebbian reverberations.

Das weltweite Ausmaß der Urbanisierung zählt zu den großen ökologischen Herausforderungen des 21. Jahrhunderts. Die Fernerkundung bietet die Möglichkeit das Verständnis dieses Prozesses und seiner Auswirkungen zu erweitern. Der Fokus dieser Arbeit lag in der Quantifizierung der städtischen Landbedeckung mittels Maschinellen Lernens und räumlich unterschiedlich aufgelöster Hyperspektraldaten. Untersuchungen berücksichtigten innovative methodische Entwicklungen und neue Möglichkeiten, die durch die bevorstehende Satellitenmission EnMAP geschaffen werden. Auf Basis von Bilder des flugzeugestützten HyMap Sensors mit Auflösungen von 3,6 m und 9 m sowie simulierten EnMAP-Daten mit einer Auflösung von 30 m wurde eine Kartierung entlang des Stadt-Umland-Gradienten Berlins durchgeführt. Im ersten Teil der Arbeit wurde die Kombination von Support Vektor Regression mit synthetischen Trainingsdaten für die Subpixelkartierung eingeführt. Ergebnisse zeigen, dass sich der Ansatz gut zur Quantifizierung thematisch relevanter und spektral komplexer Oberflächenarten eignet, dass er verbesserte Ergebnisse gegenüber weiteren Subpixelverfahren erzielt, und sich als universell einsetzbar hinsichtlich der räumlichen Auflösung erweist. Im zweiten Teil der Arbeit wurde der Wert zukünftiger EnMAP-Daten für die städtische Fernerkundung abgeschätzt. Detaillierte Untersuchungen unterstreichen deren Eignung für eine verbesserte und erweiterte Beschreibung der Stadt nach dem bewährten Vegetation-Impervious-Soil-Schema. Analysen der Möglichkeiten und Grenzen zeigen sowohl Nachteile durch die höhere Anzahl von Mischpixel im Vergleich zu hyperspektralen Flugzeugdaten als auch Vorteile aufgrund der verbesserten Differenzierung städtischer Materialien im Vergleich zu multispektralen Daten. Insgesamt veranschaulicht diese Arbeit, dass die Kombination von hyperspektraler Satellitenbildfernerkundung mit Methoden des Maschinellen Lernens eine neue Qualität in die städtische Fernerkundung bringen kann. / The global dimension of urbanization constitutes a great environmental challenge for the 21st century. Remote sensing is a valuable Earth observation tool, which helps to better understand this process and its ecological implications. The focus of this work was to quantify urban land cover by means of machine learning and imaging spectrometer data at multiple spatial scales. Experiments considered innovative methodological developments and novel opportunities in urban research that will be created by the upcoming hyperspectral satellite mission EnMAP. Airborne HyMap data at 3.6 m and 9 m resolution and simulated EnMAP data at 30 m resolution were used to map land cover along an urban-rural gradient of Berlin. In the first part of this work, the combination of support vector regression with synthetically mixed training data was introduced as sub-pixel mapping technique. Results demonstrate that the approach performs well in quantifying thematically meaningful yet spectrally challenging surface types. The method proves to be both superior to other sub-pixel mapping approaches and universally applicable with respect to changes in spatial scales. In the second part of this work, the value of future EnMAP data for urban remote sensing was evaluated. Detailed explorations on simulated data demonstrate their suitability for improving and extending the approved vegetation-impervious-soil mapping scheme. Comprehensive analyses of benefits and limitations of EnMAP data reveal both challenges caused by the high numbers of mixed pixels, when compared to hyperspectral airborne imagery, and improvements due to the greater material discrimination capability when compared to multispectral spaceborne imagery. In summary, findings demonstrate how combining spaceborne imaging spectrometry and machine learning techniques could introduce a new quality to the field of urban remote sensing.