Principal Manifold Learning by Sparse Grids

The construction of lower-dimensional manifolds from high-dimensional data is an important task in data mining, machine learning and statistics. The authors consider principal manifolds as a regularized, non-linear empirical quantization error functional.

Подобные документы

  • Multidimensional data distributions with complex topologies and variable local dimensions. A new type of low-dimensional "principal object": a principal cubic complex. The method of topological grammars with the minimization of an elastic energy.

    статья, добавлен 08.02.2013

  • Analyzing multi-dimensional data with complex geometry and to identify low-dimensional "principal objects" that relate to the optimal projection while losing the least amount of information. The methods for dimensionality reduction of microarray data.

    статья, добавлен 08.02.2013

  • The algorithm of the basic components of a linear analysis. Work on nonlinear PCA, or NLPCA, autoassociative's neural networks, principal curves and manifolds, kernel the combination's of these approaches. The problem areas that require research.

    статья, добавлен 08.02.2013

  • Auto-associative models as a new tool for building nonlinear principal component analysis methods. The successive approximations of a dataset by manifolds of increasing dimensions. The theoretical comparison between PCA and auto-associative models.

    статья, добавлен 08.02.2013

  • Linear Principal Components. A linear model formulation. The Principal Curve and Surface models. Theory for principal curves and surfaces. Algorithmic details. Estimation of curves and surfaces. Gold assay pairs. Generalized linear principal components.

    дипломная работа, добавлен 08.02.2013

  • Spatial relationships and examples of three-dimensional figures. The calculation of the surface area of three-dimensional figures and volume of cylinders and prisms. The ratio of the volume of the cone and the pyramid to the volume of cylinder and prism.

    презентация, добавлен 10.11.2015

  • Determining the extent of interpolated approximation spaces generated by regular elliptic operators on compact manifolds. Features and through the application of Jackson-Bernstein inequalities for spectral approximation regular elliptic operators.

    статья, добавлен 25.03.2016

  • Subject and method of statistical science. Elements of probability theory. Random variables and their distribution laws. Fundamental of statistical observation. Grouping, consolidated return and data presentation. Basics of averages statistics method.

    учебное пособие, добавлен 24.12.2013

  • The unsupervised methods required to reduce the dimension of the data set and to extract meaningful biological information. This work shows that Independent Component Analysis is a promising approach for the analysis of genome-wide transcriptomic data.

    статья, добавлен 08.02.2013

  • Characterization of symmetric linear functionals as the simplest polynomials. The proof of the theory that every symmetric continuous linear functional on the complex space L (0,1) can be represented as the Lebesgue integral, multiplied by a constant.

    статья, добавлен 25.03.2016

  • Меры сходства между объектами в метрическом и конкурентном пространствах. Использование функции конкурентного сходства для создания эффективных алгоритмов решения всех основных задач Data Mining, получения количественной оценки компактности образов.

    статья, добавлен 28.10.2018

  • Approximations in Scientic Computation, сomputer Arithmetic, mathematical Software. Linear Systems, solving Linear Systems, Iterative Methods for Linear Systems. Linear Least Squares, eigenvalues and Singular Values, Nonlinear Equations, optimization.

    учебное пособие, добавлен 25.11.2013

  • Bayesian hierarchical models for data analysis in this context. At the first stage of the model, survival times can he modelled via the Cox partial likelihood, using a justification due to Kalbfleisch. A questionable parametric assumptions are avoided.

    статья, добавлен 10.09.2012

  • Normed linear spaces. Contraction mapping theorem. Applications to differential and integral equations. Linear transformations. Product spaces and Fubini's theorem. Projection and self-adjoint operators. Gram-Schmidt orthonormalization. Fourier analysis.

    курс лекций, добавлен 03.07.2013

  • Using methods of discrete mathematics in the field of virtual isolation and its application in the scientific theory of numbers, groups, combinatorics and graph theory. Specifics arithmetic methods data comparisons. Construction of a matrix atom.

    книга, добавлен 08.03.2014

  • Thin and sparse metric spaces as asymptotic counterparts of discrete and very close to discrete metric spaces respectively. Classify thin metric spaces up to coarse equivalence. The types of sparse spaces and construct the spaces of distinct types.

    статья, добавлен 19.02.2016

  • Practical methods for solving nonlinear equations. Metrological support of iterative methods for solving nonlinear equations with inaccurate initial data. Modeling the behavior of iterative methods for different values of initial data and their errors.

    статья, добавлен 11.09.2022

  • Assessment of convergence result for the sequential procedure in the form of alternating maximization to the maximum likelihood estimation for a family of models - Generalized linear models. Variable Linear Regression procedure. Using techniques ALS.

    курсовая работа, добавлен 27.08.2016

  • Solving Linear Systems with the Inverse. Transposes and Symmetric Matrices. General Linear Systems. The Fundamental Matrix Subspaces. Minimization of Quadratic Functions. Computations in Orthogonal Bases. Orthogonal Polynomials and Least Squares.

    учебное пособие, добавлен 25.11.2013

  • Characteristic of a Krull–Schmidt Theorem for nonassociative algebras. Definition of the upper annihilating series. Study of some families of nilpotent evolution algebras. Classification of four- and five-dimensional nilpotent evolution algebras.

    статья, добавлен 24.01.2017

  • Conditions for nite sequences of positive numbers to be certain parts of spectra of the Dirichlet-Dirichlet, Dirichlet-Neumann, Neumann-Dirichlet and Neumann-Neumann boundary value problems generated by the same Stieltjes string recurrence relations.

    статья, добавлен 19.02.2016

  • Search of the Dirichlet series with zero abscissa of absolute convergence. Study of asymptotic equality features. The convergence at different points. The role of the Dirichlet series in number theory. The behavior of the three-dimensional function.

    статья, добавлен 19.02.2016

  • Further, we found sufficient conditions on linear Volterra integro-differential equations of the third kind to be correct in analytical functions which generalize the well-known fact of correctness of a Volterra integral equation of the second kind.

    статья, добавлен 12.08.2020

  • Finding common solutions of linear differential equations of the fourth order, the coefficients of which satisfy the system of three differential equations of the first order. The problem of conformal mapping of polygons, whose sides are arcs of circles.

    статья, добавлен 04.02.2017

  • The spectral properties of a fourth-order functional-differential operator with a summable potential are studied. The boundary conditions are separated. The solution of the functional-differential equation. The solution of the Volterra integral equation.

    статья, добавлен 14.08.2020

Работы в архивах красиво оформлены согласно требованиям ВУЗов и содержат рисунки, диаграммы, формулы и т.д.
PPT, PPTX и PDF-файлы представлены только в архивах.
Рекомендуем скачать работу и оценить ее, кликнув по соответствующей звездочке.