Hochschule Konstanz University of Applied Sciences
Hochschulschriftenserver der HTWG KonstanzNot a member yet
2319 research outputs found
Sort by
Identifikation von Schlaf- und Wachzuständen durch die Auswertung von Atem- und Bewegungssignalen
No full textNotizen zur Stadtbaukunst
No full textBedarf es heute einer institutionalisierten Stadtbaukunst? Als Gegenbewegung zur schematischen, mechanischen Stadterweiterungspraxis entwickelte sich Stadtbaukunst gegen Ende des 19. Jahrhunderts und fand in Praxis und Diskurs ihren Höhepunkt vor dem Ersten Weltkrieg. Seit einigen Jahren wird nun von verschiedenen Seiten deren Neuentdeckung und Neubewertung eingefordert. Doch die Lage heute ist eine andere als die vor über hundert Jahren. Was bedeutet das für die Idee einer Stadtbaukunst im Zusammenspiel mit Städtebau, Architektur und Stadtplanung
Transformation Models for Flexible Posteriors in Variational Bayes
No full textThe main challenge in Bayesian models is to determine the posterior for the model parameters. Already, in models with only one or few parameters, the analytical posterior can only be determined in special settings. In Bayesian neural networks, variational inference is widely used to approximate difficult-to-compute posteriors by variational distributions. Usually, Gaussians are used as variational distributions (Gaussian-VI) which limits the quality of the approximation due to their limited flexibility. Transformation models on the other hand are flexible enough to fit any distribution. Here we present transformation model-based variational inference (TM-VI) and demonstrate that it allows to accurately approximate complex posteriors in models with one parameter and also works in a mean-field fashion for multi-parameter models like neural networks
Prediction of melanoma types using semi-structured Bayesian deep learning models
Get PDFInterpretability and uncertainty modeling are important key factors for medical applications. Moreover, data in medicine are often available as a combination of unstructured data like images and structured predictors like patient’s metadata. While deep learning models are state-of-the-art for image classification, the models are often referred to as ’black-box’, caused by the lack of interpretability. Moreover, DL models are often yielding point predictions and are too confident about the parameter estimation and outcome predictions.
On the other side with statistical regression models, it is possible to obtain interpretable predictor effects and capture parameter and model uncertainty based on the Bayesian approach. In this thesis, a publicly available melanoma dataset, consisting of skin lesions and patient’s age, is used to predict the melanoma types by using a semi-structured model, while interpretable components and model uncertainty is quantified. For Bayesian models, transformation model-based variational inference (TM-VI) method is used to determine the posterior distribution of the parameter. Several model constellations consisting of patient’s age and/or skin lesion were implemented and evaluated. Predictive performance was shown to be best by using a combined model of image and patient’s age, while providing the interpretable posterior distribution of the regression coefficient is possible. In addition, integrating uncertainty in image and tabular parts results in larger variability of the outputs corresponding to high uncertainty of the single model components
Materialoberfläche wichtig für Qualität und Stahl
No full textAm Institut für Werkstoffsystemtechnik (WITG) wurden die Einflussfaktoren der Bearbeitung auf die Oberflächeneigenschaften sowie die daraus resultierende Korrosionsbeständigkeit bei nichtrostenden Stählen für die Produktion von pharmazeutischen Produkten untersucht
Joint Parameter Estimation and Trajectory Tracking of Bounding Boxes
No full textIn multi-extended object tracking, parameters (e.g., extent) and trajectory are often determined independently. In this paper, we propose a joint parameter and trajectory (JPT) state and its integration into the Bayesian framework. This allows processing measurements that contain information about parameters and states. Examples of such measurements are bounding boxes given from an image processing algorithm. It is shown that this approach can consider correlations between states and parameters. In this paper, we present the JPT Bernoulli filter. Since parameters and state elements are considered in the weighting of the measurement data assignment hypotheses, the performance is higher than with the conventional Bernoulli filter. The JPT approach can be also used for other Bayes filters
Trajectory Tracking of a Fully-actuated Surface Vessel using Nonlinear Model Predictive Control
No full textThe trajectory tracking problem for a fully-actuated real-scaled surface vessel is addressed in this paper. The unknown hydrodynamic and propulsion parameters of the vessel’s dynamic model were identified using an experimental maneuver-based identification process. Then, a nonlinear model predictive control (NMPC) scheme is designed and the controller’s performance is assessed through the variation of NMPC parameters and constraints tightening for tracking a curved trajectory
Einsatz von Bankettbeton bei schmalen und stark beanspruchten Ortsverbindungs- und Kreisstraßen
No full textRechtsrahmen eines Start-up-Managements
No full textDieser Ratgeber orientiert sich bei der Vorstellung der rechtlichen Rahmenbedingungen am Ablauf des Gründungsprozesses eines Start-ups:
- Anmeldung,
- Schutz der Geschäftsidee,
- Wahl der Rechtsform und
- Marketingaktivitäten