Approximate tangent space

Last updated

In geometric measure theory an approximate tangent space is a measure theoretic generalization of the concept of a tangent space for a differentiable manifold.

Contents

Definition

In differential geometry the defining characteristic of a tangent space is that it approximates the smooth manifold to first order near the point of tangency. Equivalently, if we zoom in more and more at the point of tangency the manifold appears to become more and more straight, asymptotically tending to approach the tangent space. This turns out to be the correct point of view in geometric measure theory.

Definition for sets

Definition. Let be a set that is measurable with respect to m-dimensional Hausdorff measure , and such that the restriction measure is a Radon measure. We say that an m-dimensional subspace is the approximate tangent space to at a certain point , denoted , if

as

in the sense of Radon measures. Here for any measure we denote by the rescaled and translated measure:

Certainly any classical tangent space to a smooth submanifold is an approximate tangent space, but the converse is not necessarily true.

Multiplicities

The parabola

is a smooth 1-dimensional submanifold. Its tangent space at the origin is the horizontal line . On the other hand, if we incorporate the reflection along the x-axis:

then is no longer a smooth 1-dimensional submanifold, and there is no classical tangent space at the origin. On the other hand, by zooming in at the origin the set is approximately equal to two straight lines that overlap in the limit. It would be reasonable to say it has an approximate tangent space with multiplicity two.

Definition for measures

One can generalize the previous definition and proceed to define approximate tangent spaces for certain Radon measures, allowing for multiplicities as explained in the section above.

Definition. Let be a Radon measure on . We say that an m-dimensional subspace is the approximate tangent space to at a point with multiplicity , denoted with multiplicity , if

as

in the sense of Radon measures. The right-hand side is a constant multiple of m-dimensional Hausdorff measure restricted to .

This definition generalizes the one for sets as one can see by taking for any as in that section. It also accounts for the reflected paraboloid example above because for we have with multiplicity two.

Relation to rectifiable sets

The notion of approximate tangent spaces is very closely related to that of rectifiable sets. Loosely speaking, rectifiable sets are precisely those for which approximate tangent spaces exist almost everywhere. The following lemma encapsulates this relationship:

Lemma. Let be measurable with respect to m-dimensional Hausdorff measure. Then is m-rectifiable if and only if there exists a positive locally -integrable function such that the Radon measure

has approximate tangent spaces for -almost every .

Related Research Articles

In mathematics, specifically in measure theory, a Borel measure on a topological space is a measure that is defined on all open sets. Some authors require additional restrictions on the measure, as described below.

In statistics, maximum likelihood estimation (MLE) is a method of estimating the parameters of an assumed probability distribution, given some observed data. This is achieved by maximizing a likelihood function so that, under the assumed statistical model, the observed data is most probable. The point in the parameter space that maximizes the likelihood function is called the maximum likelihood estimate. The logic of maximum likelihood is both intuitive and flexible, and as such the method has become a dominant means of statistical inference.

In mathematical analysis, Hölder's inequality, named after Otto Hölder, is a fundamental inequality between integrals and an indispensable tool for the study of Lp spaces.

In calculus, absolute continuity is a smoothness property of functions that is stronger than continuity and uniform continuity. The notion of absolute continuity allows one to obtain generalizations of the relationship between the two central operations of calculus—differentiation and integration. This relationship is commonly characterized in the framework of Riemann integration, but with absolute continuity it may be formulated in terms of Lebesgue integration. For real-valued functions on the real line, two interrelated notions appear: absolute continuity of functions and absolute continuity of measures. These two notions are generalized in different directions. The usual derivative of a function is related to the Radon–Nikodym derivative, or density, of a measure. We have the following chains of inclusions for functions over a compact subset of the real line:

In mathematics, a Killing vector field, named after Wilhelm Killing, is a vector field on a Riemannian manifold that preserves the metric. Killing fields are the infinitesimal generators of isometries; that is, flows generated by Killing fields are continuous isometries of the manifold. More simply, the flow generates a symmetry, in the sense that moving each point of an object the same distance in the direction of the Killing vector will not distort distances on the object.

In theoretical physics, there are many theories with supersymmetry (SUSY) which also have internal gauge symmetries. Supersymmetric gauge theory generalizes this notion.

In statistics, a parametric model or parametric family or finite-dimensional model is a particular class of statistical models. Specifically, a parametric model is a family of probability distributions that has a finite number of parameters.

In mathematics, more particularly in functional analysis, differential topology, and geometric measure theory, a k-current in the sense of Georges de Rham is a functional on the space of compactly supported differential k-forms, on a smooth manifold M. Currents formally behave like Schwartz distributions on a space of differential forms, but in a geometric setting, they can represent integration over a submanifold, generalizing the Dirac delta function, or more generally even directional derivatives of delta functions (multipoles) spread out along subsets of M.

<span class="mw-page-title-main">Complex torus</span>

In mathematics, a complex torus is a particular kind of complex manifold M whose underlying smooth manifold is a torus in the usual sense. Here N must be the even number 2n, where n is the complex dimension of M.

In mathematics, a π-system on a set is a collection of certain subsets of such that

In mathematics, the Weyl character formula in representation theory describes the characters of irreducible representations of compact Lie groups in terms of their highest weights. It was proved by Hermann Weyl. There is a closely related formula for the character of an irreducible representation of a semisimple Lie algebra. In Weyl's approach to the representation theory of connected compact Lie groups, the proof of the character formula is a key step in proving that every dominant integral element actually arises as the highest weight of some irreducible representation. Important consequences of the character formula are the Weyl dimension formula and the Kostant multiplicity formula.

In mathematics, Gaussian measure is a Borel measure on finite-dimensional Euclidean space Rn, closely related to the normal distribution in statistics. There is also a generalization to infinite-dimensional spaces. Gaussian measures are named after the German mathematician Carl Friedrich Gauss. One reason why Gaussian measures are so ubiquitous in probability theory is the central limit theorem. Loosely speaking, it states that if a random variable X is obtained by summing a large number N of independent random variables of order 1, then X is of order and its law is approximately Gaussian.

In mathematics, there is a folklore claim that there is no analogue of Lebesgue measure on an infinite-dimensional Banach space. The theorem this refers to states that there is no translationally invariant measure on a separable Banach space - because if any ball has nonzero non-infinite volume, a slightly smaller ball has zero volume, and countable many such smaller balls cover the space. The folklore statement, however, is entirely false. The countable product of Lebesgue measure is translationally invariant and gives the intuitive notion of volume as the infinite product of lengths, only the domain on which this product measure is defined must necessarily be non-separable, and the measure itself is not sigma finite.

<span class="mw-page-title-main">Autoencoder</span> Neural network that learns efficient data encoding in an unsupervised manner

An autoencoder is a type of artificial neural network used to learn efficient codings of unlabeled data. An autoencoder learns two functions: an encoding function that transforms the input data, and a decoding function that recreates the input data from the encoded representation. The autoencoder learns an efficient representation (encoding) for a set of data, typically for dimensionality reduction.

Covariance matrix adaptation evolution strategy (CMA-ES) is a particular kind of strategy for numerical optimization. Evolution strategies (ES) are stochastic, derivative-free methods for numerical optimization of non-linear or non-convex continuous optimization problems. They belong to the class of evolutionary algorithms and evolutionary computation. An evolutionary algorithm is broadly based on the principle of biological evolution, namely the repeated interplay of variation and selection: in each generation (iteration) new individuals are generated by variation, usually in a stochastic way, of the current parental individuals. Then, some individuals are selected to become the parents in the next generation based on their fitness or objective function value . Like this, over the generation sequence, individuals with better and better -values are generated.

In measure theory, a field of mathematics, the Hausdorff density measures how concentrated a Radon measure is at some point.

In measure theory, tangent measures are used to study the local behavior of Radon measures, in much the same way as tangent spaces are used to study the local behavior of differentiable manifolds. Tangent measures are a useful tool in geometric measure theory. For example, they are used in proving Marstrand's theorem and Preiss' theorem.

In mathematics, the spectral theory of ordinary differential equations is the part of spectral theory concerned with the determination of the spectrum and eigenfunction expansion associated with a linear ordinary differential equation. In his dissertation, Hermann Weyl generalized the classical Sturm–Liouville theory on a finite closed interval to second order differential operators with singularities at the endpoints of the interval, possibly semi-infinite or infinite. Unlike the classical case, the spectrum may no longer consist of just a countable set of eigenvalues, but may also contain a continuous part. In this case the eigenfunction expansion involves an integral over the continuous part with respect to a spectral measure, given by the Titchmarsh–Kodaira formula. The theory was put in its final simplified form for singular differential equations of even degree by Kodaira and others, using von Neumann's spectral theorem. It has had important applications in quantum mechanics, operator theory and harmonic analysis on semisimple Lie groups.

In mathematics, the Poisson boundary is a measure space associated to a random walk. It is an object designed to encode the asymptotic behaviour of the random walk, i.e. how trajectories diverge when the number of steps goes to infinity. Despite being called a boundary it is in general a purely measure-theoretical object and not a boundary in the topological sense. However, in the case where the random walk is on a topological space the Poisson boundary can be related to the Martin boundary which is an analytic construction yielding a genuine topological boundary. Both boundaries are related to harmonic functions on the space via generalisations of the Poisson formula.

In physics, the poppy-seed bagel theorem concerns interacting particles confined to a bounded surface when the particles repel each other pairwise with a magnitude that is proportional to the inverse distance between them raised to some positive power . In particular, this includes the Coulomb law observed in Electrostatics and Riesz potentials extensively studied in Potential theory. For such particles, a stable equilibrium state, which depends on the parameter , is attained when the associated potential energy of the system is minimal. For large numbers of points, these equilibrium configurations provide a discretization of which may or may not be nearly uniform with respect to the surface area of . The poppy-seed bagel theorem asserts that for a large class of sets , the uniformity property holds when the parameter is larger than or equal to the dimension of the set . For example, when the points are confined to the 2-dimensional surface of a torus embedded in 3 dimensions, one can create a large number of points that are nearly uniformly spread on the surface by imposing a repulsion proportional to the inverse square distance between the points, or any stronger repulsion. From a culinary perspective, to create the nearly perfect poppy-seed bagel where bites of equal size anywhere on the bagel would contain essentially the same number of poppy seeds, impose at least an inverse square distance repelling force on the seeds.

References