Kaiserslautern - Fachbereich Mathematik
Refine
Year of publication
- 2007 (34) (remove)
Document Type
- Doctoral Thesis (14)
- Preprint (14)
- Diploma Thesis (3)
- Report (3)
Has Fulltext
- yes (34)
Keywords
- Elastoplastizität (2)
- Mixture Models (2)
- Multiresolution Analysis (2)
- Optionspreistheorie (2)
- Sobolev spaces (2)
- Spline-Approximation (2)
- localizing basis (2)
- 2-d kernel regression (1)
- A-infinity-bimodule (1)
- A-infinity-category (1)
- A-infinity-functor (1)
- Adjazenz-Beziehungen (1)
- Algorithmics (1)
- Asympotic Analysis (1)
- Asymptotic Analysis (1)
- Asymptotic Expansion (1)
- Asymptotische Entwicklung (1)
- Biorthogonalisation (1)
- Bondindizes (1)
- Continuum mechanics (1)
- Earth's disturbing potential (1)
- Elasticity (1)
- Elastizität (1)
- Elastoplasticity (1)
- Endliche Geometrie (1)
- Finanzmathematik (1)
- Gauge Distances (1)
- Geometric Ergodicity (1)
- Geostrophic flow (1)
- Granular flow (1)
- Hamilton-Jacobi-Differentialgleichung (1)
- Harmonische Spline-Funktion (1)
- Helmholtz Type Boundary Value Problems (1)
- Homologische Algebra (1)
- Homotopiehochhebungen (1)
- Homotopy lifting (1)
- Hysterese (1)
- Identifiability (1)
- Injectivity of mappings (1)
- Injektivität von Abbildungen (1)
- Internationale Diversifikation (1)
- Inverses Problem (1)
- Jiang's constitutive model (1)
- Jiangsches konstitutives Gesetz (1)
- Kategorientheorie (1)
- Knapsack (1)
- Kombinatorische Optimierung (1)
- Kontinuumsmechanik (1)
- Legendre Wavelets (1)
- Lineare Integralgleichung (1)
- Locally Supported Zonal Kernels (1)
- Location Theory (1)
- Machine Scheduling (1)
- Markov Chain (1)
- Mehrdimensionale Spline-Funktion (1)
- Minimal spannender Baum (1)
- NP (1)
- Neural networks (1)
- Nichtlineare/große Verformungen (1)
- Nonlinear/large deformations (1)
- Nonparametric AR-ARCH (1)
- Optimale Portfolios (1)
- Optimization (1)
- Ovoid (1)
- Portfoliomanagement (1)
- Quadratischer Raum (1)
- Radial Basis Functions (1)
- Randwertproblem (1)
- Ratenunabhängigkeit (1)
- Seismische Tomographie (1)
- Serre functor (1)
- Sobolevräume (1)
- Spherical Fast Wavelet Transform (1)
- Spherical Wavelets (1)
- Sphärische Approximation (1)
- Spline-Interpolation (1)
- Sprung-Diffusions-Prozesse (1)
- Stochastische Zinsen (1)
- Stochastische dynamische Optimierung (1)
- Stop- und Spieloperator (1)
- Time-Space Multiresolution Analysis (1)
- Transaktionskosten (1)
- Variationsungleichungen (1)
- Vectorfield approximation (1)
- Vektorfeldapproximation (1)
- Volatilität (1)
- Wave Based Method (1)
- adjacency relations (1)
- algorithm (1)
- bocses (1)
- boundary value problem (1)
- combinatorial optimization (1)
- conditional quantile (1)
- consistency (1)
- convergence (1)
- cut (1)
- cut basis problem (1)
- data structure (1)
- data-adaptive bandwidth choice (1)
- deflections of the vertical (1)
- degenerations of an elliptic curve (1)
- density gradient equation (1)
- elastoplasticity (1)
- finite volume method (1)
- fluid structure (1)
- fundamental cut (1)
- graph and network algorithm (1)
- heuristic (1)
- hub location (1)
- hyperbolic systems (1)
- hysteresis (1)
- image analysis (1)
- image denoising (1)
- integer programming (1)
- interface problem (1)
- inverse problem (1)
- inverse problems (1)
- iterative bandwidth choice (1)
- knapsack (1)
- lattice Boltzmann (1)
- local approximation of sea surface topography (1)
- local bandwidths (1)
- local multiscale (1)
- localization (1)
- locally supported (Green's) vector wavelets (1)
- locally supported (Green’s) vector wavelets (1)
- lokalisierende Basis (1)
- matrix problems (1)
- minimal spanning tree (1)
- minimum cost flows (1)
- minimum fundamental cut basis (1)
- multicategory (1)
- multigrid method (1)
- network flows (1)
- neural network (1)
- nonlinear diffusion filtering (1)
- nonlinear finite element method (1)
- numerics (1)
- penalization (1)
- polyhedral analysis (1)
- poroelasticity (1)
- qualitative threshold model (1)
- rate-independency (1)
- regularization (1)
- seismic tomography (1)
- sieve estimate (1)
- smoothing (1)
- spherical approximation (1)
- splines (1)
- stop- and play-operator (1)
- time series (1)
- topological asymptotic expansion (1)
- uncapacitated facility location (1)
- variational inequalities (1)
- vector bundles (1)
Faculty / Organisational entity
As a first approximation the Earth is a sphere; as a second approximation it may be considered an ellipsoid of revolution. The deviations of the actual Earth's gravity field from the ellipsoidal 'normal' field are so small that they can be understood to be linear. The splitting of the Earth's gravity field into a 'normal' and a remaining small 'disturbing' field considerably simplifies the problem of its determination. Under the assumption of an ellipsoidal Earth model high observational accuracy is achievable only if the deviation (deflection of the vertical) of the physical plumb line, to which measurements refer, from the ellipsoidal normal is not ignored. Hence, the determination of the disturbing potential from known deflections of the vertical is a central problem of physical geodesy. In this paper we propose a new, well-promising method for modelling the disturbing potential locally from the deflections of the vertical. Essential tools are integral formulae on the sphere based on Green's function of the Beltrami operator. The determination of the disturbing potential from deflections of the vertical is formulated as a multiscale procedure involving scale-dependent regularized versions of the surface gradient of the Green function. The modelling process is based on a multiscale framework by use of locally supported surface curl-free vector wavelets.
The nowadays increasing number of fields where large quantities of data are collected generates an emergent demand for methods for extracting relevant information from huge databases. Amongst the various existing data mining models, decision trees are widely used since they represent a good trade-off between accuracy and interpretability. However, one of their main problems is that they are very instable, which complicates the process of the knowledge discovery because the users are disturbed by the different decision trees generated from almost the same input learning samples. In the current work, binary tree classifiers are analyzed and partially improved. The analysis of tree classifiers goes from their topology from the graph theory point of view to the creation of a new tree classification model by means of combining decision trees and soft comparison operators (Mlynski, 2003) with the purpose to not only overcome the well known instability problem of decision trees, but also in order to confer the ability of dealing with uncertainty. In order to study and compare the structural stability of tree classifiers, we propose an instability coefficient which is based on the notion of Lipschitz continuity and offer a metric to measure the proximity between decision trees. This thesis converges towards its main part with the presentation of our model ``Soft Operators Decision Tree\'\' (SODT). Mainly, we describe its construction, application and the consistency of the mathematical formulation behind this. Finally we show the results of the implementation of SODT and compare numerically the stability and accuracy of a SODT and a crisp DT. The numerical simulations support the stability hypothesis and a smaller tendency to overfitting the training data with SODT than with crisp DT is observed. A further aspect of this inclusion of soft operators is that we choose them in a way so that the resulting goodness function (used by this method) is differentiable and thus allows to calculate the best split points by means of gradient descent methods. The main drawback of SODT is the incorporation of the unpreciseness factor, which increases the complexity of the algorithm.
The thesis is concerned with multiscale approximation by means of radial basis functions on hierarchically structured spherical grids. A new approach is proposed to construct a biorthogonal system of locally supported zonal functions. By use of this biorthogonal system of locally supported zonal functions, a spherical fast wavelet transform (SFWT) is established. Finally, based on the wavelet analysis, geophysically and geodetically relevant problems involving rotation-invariant pseudodifferential operators are shown to be efficiently and economically solvable.
This paper presents a wavelet analysis of temporal and spatial variations of the Earth's gravitational potential based on tensor product wavelets. The time--space wavelet concept is realized by combining Legendre wavelets for the time domain and spherical wavelets for the space domain. In consequence, a multiresolution analysis for both, temporal and spatial resolution, is formulated within a unified concept. The method is then numerically realized by using first synthetically generated data and, finally, several real data sets.