Probabilistic Fréchet means for time varying persistence diagrams

© 2015, Institute of Mathematical Statistics. All rights reserved.In order to use persistence diagrams as a true statistical tool, it would be very useful to have a good notion of mean and variance for a set of diagrams. In [23], Mileyko and his collaborators made the first study of the properties of the Fréchet mean in (Dp, Wp), the space of persistence diagrams equipped with the p-th Wasserstein metric. In particular, they showed that the Fréchet mean of a finite set of diagrams always exists, but is not necessarily unique. The means of a continuously-varying set of diagrams do not themselves (necessarily) vary continuously, which presents obvious problems when trying to extend the Fréchet mean definition to the realm of time-varying persistence diagrams, better known as vineyards. We fix this problem by altering the original definition of Fréchet mean so that it now becomes a probability measure on the set of persistence diagrams; in a nutshell, the mean of a set of diagrams will be a weighted sum of atomic measures, where each atom is itself a persistence diagram determined using a perturbation of the input diagrams. This definition gives for each N a map (Dp)^N→ℙ(Dp). We show that this map is Hölder continuous on finite diagrams and thus can be used to build a useful statistic on vineyards.

PIANIST-COMPOSERS WHOSE WORKS INCLUDE PIANISTIC INNOVATIONS

Throughout the piano’s history, certain composers have created innovations in the areas of virtuosity and sonority. These innovations came not only from the composers’ imagination, but also from the development of instruments and changes in musical style from one period to another. To investigate what kinds of innovations these pianist composers made, I divided them into technique and sound from Mozart to Cowell. I chose two-piano music (Sonata in D major, K.448 by Mozart and Rachmaninoff’s Second Suite) to demonstrate their experiments with varieties of textures and sonorities, using different registers of the two pianos orchestrally. En Blanc et noir by Debussy shows this composer’s deep interest and originality in piano sonorities. For solo piano music, Beethoven’s Piano Sonata Op.53 shows extensive technical invention. His use of long pedal effects shows a pianistic possibility not explored by Mozart. Hummel’s Piano Sonata in D major represents orchestral devices as well as pianistic techniques showing recent developments in the instrument. Chopin’s Ballade No.3 and Scherzo No.3 show virtuosic moments and also the expanded range of the keyboard. His Nocturne Op.27, no.2, with its sonorities resulting from the combination of pedal, and widespread accompaniments derived from Alberti bass figures, is a perfect example of Chopin’s characteristic sound-world. “Vallée d’Obermann” by Liszt uses many virtuosic techniques as well as the extreme wide ranges of keyboard in both hands to create dramatic contrasts of texture. Debussy’s etude, “Pour les Sonorités opposés” is probably the first etude designed for sonority rather than for keyboard virtuosity. Albeniz’s “Evocación” and “Triana” show Spanish atmosphere. Prokofiev’s Sonata no.3 shows frequent motoric driving elements that demand percussive virtuosity. Cowell’s piano music is some of the earliest to explore the sonorities of tone clusters and playing on the strings. This performance dissertation consists of three recitals performed in the Orchestra Room, Leah Smith Hall, and Gildenhorn Recital Hall at the University of Maryland, College Park. These recitals are documented on compact disc recordings that are housed within the University of Maryland Library System.

Persistent Homology Analysis of Brain Artery Trees.

New representations of tree-structured data objects, using ideas from topological data analysis, enable improved statistical analyses of a population of brain artery trees. A number of representations of each data tree arise from persistence diagrams that quantify branching and looping of vessels at multiple scales. Novel approaches to the statistical analysis, through various summaries of the persistence diagrams, lead to heightened correlations with covariates such as age and sex, relative to earlier analyses of this data set. The correlation with age continues to be significant even after controlling for correlations from earlier significant summaries.

Universal elements for non-linear operators and their applications

We prove that under certain topological conditions on the set of universal elements of a continuous map T acting on a topological space X, that the direct sum T and M_g is universal, where M_g is multiplication by a generating element of a compact topological group. We use this result to characterize R_+-supercyclic operators and to show that whenever T is a supercyclic operator and z_1,...,z_n are pairwise different non-zero complex numbers, then the operator z_1T\oplus ... \oplus z_n T is cyclic. The latter answers affirmatively a question of Bayart and Matheron.

Multiplicative spectra of Banach spaces

The multiplicative spectrum of a complex Banach space X is the class K(X) of all (automatically compact and Hausdorff) topological spaces appearing as spectra of Banach algebras (X,*) for all possible continuous multiplications on X turning X into a commutative associative complex algebra with the unity. The properties of the multiplicative spectrum are studied. In particular, we show that K(X^n) consists of countable compact spaces with at most n non-isolated points for any separable hereditarily indecomposable Banach space X. We prove that K(C[0,1]) coincides with the class of all metrizable compact spaces.

On weak and strong Peano theorems

We say that the Peano theorem holds for a topological vector space $E$ if, for any continuous mapping $f : {\Bbb R}\times E \to E$ and any $(t(0), x(0))$ is an element of ${\Bbb R}\times E$, the Cauchy problem $\dot x(t) = f(t,x(t))$, $x(t(0)) = x(0)$, has a solution in some neighborhood of $t(0)$. We say that the weak version of Peano theorem holds for $E$ if, for any continuous map $f : {\Bbb R}\times E \to E$, the equation $\dot x(t) = f (t, x(t))$ has a solution on some interval. We construct an example (answering a question posed by S. G. Lobanov) of a Hausdorff locally convex topological vector space E for which the weak version of Peano theorem holds and the Peano theorem fails to hold. We also construct a Hausdorff locally convex topological vector space E for which the Peano theorem holds and any barrel in E is neither compact nor sequentially compact.

On linear extension operators

We construct a countable-dimensional Hausdorff locally convex topological vector space $E$ and a stratifiable closed linear subspace $F$ subset of $E$ such that any linear extension operator from $C_b(F)$ to $C_b(E)$ is unbounded (here $C_b(X)$ stands for the Banach space of continuous bounded real-valued functions on $X$).

Some results on solvability of ordinary linear differential equations in locally convex spaces

Let $\Gamma$ be the class of sequentially complete locally convex spaces such that an existence theorem holds for the linear Cauchy problem $\dot x = Ax$, $x(0) = x_0$ with respect to functions $x: R\to E$. It is proved that if $E\in \Gamma$, then $E\times R^A$ is-an-element-of $\Gamma$ for an arbitrary set $A$. It is also proved that a topological product of infinitely many infinite-dimensional Frechet spaces, each not isomorphic to $\omega$, does not belong to $\Gamma$.

K-Theory of non-linear projective toric varieties

We define a category of quasi-coherent sheaves of topological spaces on projective toric varieties and prove a splitting result for its algebraic K-theory, generalising earlier results for projective spaces. The splitting is expressed in terms of the number of interior lattice points of dilations of a polytope associated to the variety. The proof uses combinatorial and geometrical results on polytopal complexes. The same methods also give an elementary explicit calculation of the cohomology groups of a projective toric variety over any commutative ring.

Del Fichero supremo a la Biblioteca de Babel: Salaverría, precursor de Borges

A juzgar por alguna declaración más bien negativa de Borges sobre la literatura española, se podría creer que esta no influyó demasiado en él. Sin embargo, existen indicios de que pudo haberse inspirado también en determinados escritores españoles coetáneos hoy casi olvidados. Uno de ellos pudo ser José María Salaverría, entre cuyos relatos destaca “El fichero supremo” (1926), del que se ha dicho que “anticipa algunas de las preocupaciones características de un tipo de relato que Jorge Luis Borges elevará años después a la máxima categoría estética”. De hecho, recuerda a “La biblioteca de Babel” (1941) borgiana por su planteamiento hasta el punto de que podría pensarse que el maestro argentino pudo tener presente, a la hora de escribir esa obra maestra, ese cuento de Salaverría, el cual se publicó por primera vez en Caras y Caretas, una revista porteña que Borges reconoció “devorar” en su juventud. Sin embargo, el interés mayor de la comparación entre “El fichero supremo” y “La biblioteca de Babel” no radica tanto en el carácter de posible fuente del primero como en el contraste entre sus formas de presentación narrativa: desde fuera y en tercera persona en Salaverría, en un marco realista; y desde dentro y en primera persona, prácticamente sin marco, en Borges. Este parece desarrollar, en el registro propio de la “imaginación razonada” descrito por él mismo, una virtualidad presente en el relato de Salaverría, cuya comparación con “La biblioteca de Babel” puede suscitar también alguna reflexión sobre el enigma de la identidad y el carácter de la voz enunciadora de la biblioteca universal de Babel. Al menos, esta parece haber hecho realidad en cierto modo, de forma sublime, el patético sueño divino del archivero imaginado por Salaverría.

Neuropeptide-like protein diversity in phylum Nematoda

This study reports the identification of nematode neuropeptide-like protein (nlp) sequelogs from the GenBank expressed sequence tag (EST) database, using BLAST (Basic Local Alignment Search Tool) search methodology. Search strings derived from peptides encoded by the 45 known Caenorhabatitis elegans nlp genes were used to identify more than 1000 ESTs encoding a total of 26 multi-species nlp sequelogs. The remaining 18 nlps (nlp-4, -16, -24 through -36, -39, -41 and -45) were identified only in C elegans, while the sole EST representative of nlp-23 was from Caenorhabditis remanei. Several ESTs encoding putative antibacterial peptides similar to those encoded by the C elegans genes nlp-24-33 were observed in several parasite species. A novel gene (nlp-46) was identified, encoding a single, amidated dodecapeptide (NIA[I/T]GR[G/A]DG[F/L]RPG) in eight species. Secretory signal peptides were identified in at least one species representing each nlp sequelog, confirming that all 46 nematode nlp genes encode secretory peptides. A random sub-set of C elegans NLPs was tested physiologically in Ascaris suum ovijector and body wall muscle bioassays. None of the peptides tested were able to modulate ovijector activity, while only three displayed measurable myoactivity on somatic body wall muscle. AFAAGWNRamide (from nlp-23) and AVNPFLDSIamide (nlp-3) both produced a relaxation of body wall muscle, while AIPFNGGMYamide (nlp-10) induced a transient contraction. Numerical analyses of nip-encoding ESTs demonstrate that nlp-3, -13, -14, -15 and -18 are amongst the most highly represented transcripts in the dataset. Using available bioinformatics resources, this study delineates the nlp complement of phylum Nematoda, providing a rich source of neuropeptide ligands for deorphanisation of nematode neuropeptide receptors. (C) 2008 Australian Society for Parasitology Inc. Published by Elsevier Ltd. All rights reserved.

Exploratory analysis of spatiotemporal patterns of cellular automata by clustering compressibility

In this paper we study the classification of spatiotemporal pattern of one-dimensional cellular automata (CA) whereas the classification comprises CA rules including their initial conditions. We propose an exploratory analysis method based on the normalized compression distance (NCD) of spatiotemporal patterns which is used as dissimilarity measure for a hierarchical clustering. Our approach is different with respect to the following points. First, the classification of spatiotemporal pattern is comparative because the NCD evaluates explicitly the difference of compressibility among two objects, e.g., strings corresponding to spatiotemporal patterns. This is in contrast to all other measures applied so far in a similar context because they are essentially univariate. Second, Kolmogorov complexity, which underlies the NCD, was used in the classification of CA with respect to their spatiotemporal pattern. Third, our method is semiautomatic allowing us to investigate hundreds or thousands of CA rules or initial conditions simultaneously to gain insights into their organizational structure. Our numerical results are not only plausible confirming previous classification attempts but also shed light on the intricate influence of random initial conditions on the classification results.

Domain Bundle Boundaries in Single Crystal BaTiO3 Lamellae: Searching for Naturally Forming Dipole Flux-Closure / Quadrupole Chains

Naturally occurring boundaries between bundles of 90o stripe domains, which form in BaTiO3 lamellae on cooling through the Curie Temperature, have been characterised using both piezoresponse force microscopy (PFM) and scanning transmission electron microscopy (STEM). Detailed interpretation of the dipole configurations present at these boundaries (using data taken from PFM) shows that, in the vast majority of cases, they are composed of simple zigzag 180° domain walls. Topological information from STEM shows that, occasionally, domain bundle boundaries can support chains of dipole flux closure and quadrupole nanostructures, but these kinds of boundaries are comparatively rare; when such chains do exist, it is notable that singularities at the cores of the dipole structures are avoided. The symmetry of the boundary shows that diads and centres of inversion exist at positions where core singularities should have been expected.

Genetic Algorithm Search for Predictive Patterns in Multidimensional Time Series

Based on an algorithm for pattern matching in character strings, we implement a pattern matching machine that searches for occurrences of patterns in multidimensional time series. Before the search process takes place, time series are encoded in user-designed alphabets. The patterns, on the other hand, are formulated as regular expressions that are composed of letters from these alphabets and operators. Furthermore, we develop a genetic algorithm to breed patterns that maximize a user-defined fitness function. In an application to financial data, we show that patterns bred to predict high exchange rates volatility in training samples retain statistically significant predictive power in validation samples.