7. The Stability Theorem
|
|
- Christopher Lawson
- 5 years ago
- Views:
Transcription
1 Vin de Silva Pomona College CBMS Lecture Series, Macalester College, June 2017
2 The Persistence Stability Theorem Stability Theorem (Cohen-Steiner, Edelsbrunner, Harer 2007) The following transformation is Lipschitz: data / persistence diagram / / / / Small changes in the data lead to small changes in the diagram. Factorization data persistence module persistence diagram Geometric Stability data persistence module is Lipschitz (in various settings) Algebraic Stability (Chazal, Cohen-Steiner, Glisse, Guibas, Oudot 2009) persistence module persistence diagram is 1-Lipschitz
3 Data Simplex stream Let S be a finite simplicial complex. Let f : S! R be a function with the property that f 1 ( 1, t] is a subcomplex of S for every t 2 R. Processing Run the persistence algorithm on the sequence of simplices ordered by f -value. Metric Two such functions f, g : S! R are compared by the supremum norm: kf gk 1 =max f ( ) g( ) 2S
4 Data Sublevelset filtration Let M be a compact smooth manifold. Let f : M! R be a Morse function (critical points nondegenerate) Processing Compute the persistence diagram of H f 1 ( 1, a 0] /... / H f 1 ( 1, a n] where a 0 < < a n are the critical values. Metric Two such functions f, g : M! R are compared by the supremum norm: kf gk 1 =max f (x) g(x) x2m
5 Data Sublevelset filtration Let P be a compact polyhedron (, realization of a simplicial complex). Let f : P! R be a piecewise-linear map. Processing Compute the persistence diagram of H f 1 ( 1, a 0] /... / H f 1 ( 1, a n] where a 0 < < a n are the critical values. Metric Two such functions f, g : P! R are compared by the supremum norm: kf gk 1 =max f (x) g(x) x2p
6 Data Metric data (fixed set) Let X be a finite set. Let d : X X! R be a metric. Processing Compute the persistence diagram of H VR((X, d), r 0) /... / H VR((X, d), r n) where r 0 < < r n are the values taken by the metric. Metric Two such metrics d, e : X X! R are compared by the supremum norm: kd ek 1 =max d(x, y) x,y2p e(x, y)
7 Data Metric data (subsets of a fixed space) Let (Z, d) beametricspace. Let X Z be a finite subset. Processing Compute the persistence diagram of H VR((X, d), r 0) /... / H VR((X, d), r n) where r 0 < < r n are the values taken by the metric restricted to X. Metric Two such subsets X, Y Z are compared by the Hausdor distance: d H(X, Y )=maxapple max d(x, Y ), max d(y, X ) x2x y2y
8 Data Metric data (general) Let (X, d) beafinitemetricspace. Processing Compute the persistence diagram of H VR((X, d), r 0) /... / H VR((X, d), r n) where r 0 < < r n are the values taken by the metric. Metric Two metric spaces (X, d), (Y, e) are compared using the Gromov Hausdor distance: d GH((X, d), (Y, e)) = inf Z,f,g dh(f (X ), g(y )) The min is taken over all metric spaces Z and isometric embeddings f : X! Z, g : Y! Z.
9 Persistence modules Persistence modules indexed by R (Chazal et al. 2009) For the stability theorem, we need to work with real index values. A persistence module indexed by R is specified by the following data: AvectorspaceV t for every t 2 R. Alinearmapvt s : V s! V t whenever s apple t. We require vt t =1 Vt for all t, andvuv t t s = vu s whenever s apple t apple u. Henceforth, all persistence modules will be indexed by R unless stated otherwise.
10 Persistence modules Persistence modules indexed by R (Chazal et al. 2009) For the stability theorem, we need to work with real index values. A persistence module indexed by R is specified by the following data: AvectorspaceV t for every t 2 R. Alinearmapvt s : V s! V t whenever s apple t. We require vt t =1 Vt for all t, andvuv t t s = vu s whenever s apple t apple u. Henceforth, all persistence modules will be indexed by R unless stated otherwise. Tameness Apersistencemoduleisoffinite type if there is a finite set of critical values a 0 < a 1 < < a n such that V t = 0 for all t < a 0;and V t is constant, and finite dimensional, over each interval [a k, a k+1 ). (Here we take a n+1 =+1.)
11 Persistence modules Persistence modules indexed by R (Chazal et al. 2009) For the stability theorem, we need to work with real index values. A persistence module indexed by R is specified by the following data: AvectorspaceV t for every t 2 R. Alinearmapvt s : V s! V t whenever s apple t. We require vt t =1 Vt for all t, andvuv t t s = vu s whenever s apple t apple u. Henceforth, all persistence modules will be indexed by R unless stated otherwise. Tameness Apersistencemoduleisq-tame if vt s has finite rank whenever s < t. Fact q-tame persistence modules have well-behaved persistence diagrams.
12 ! " < / < = / Persistence modules The interleaving distance (Chazal et al. 2009) An r-interleaving between persistence modules V, W is a collection of maps t : V t! W t+r, t : W t! V t+r such that the diagrams v s v t t+2r V s / t V t / s W s+r t w s+r t+r / W t+r t+r V t+2r V s+r v s+r t+r W s w s t s W t t / V t+r w t t+2r t+r " W t+2r commute for all s < t. A0-interleavingisanisomorphismofpersistencemodules. The interleaving distance between V, W is d I(V, W )=inf r 0 r there exists an r-interleaving between V, W
13 " < < = " Data! Persistence modules Simplex streams Let S be a simplicial complex and let f, g : S! R define simplex streams. Suppose kf gk 1 apple r. Writing [f ] t = f 1 ( 1, t], [g] t = g 1 ( 1, t], we have inclusions [f ] t [g] t+r, [g] t [f ] t+r. More generally, we have diagrams [f ] s / [f ] t / [f ] t+2r [f ] s+r / [f ] t+r [g] s+r / [g] t+r [g] s / [g] t / [g] t+2r which automatically commute, since all arrows are inclusion maps. Applying homology H,weobtainthemaps t, t and required relations for an r-interleaving between the persistence modules for the two simplex streams.
14 " < < = " Data! Persistence modules Sublevelset filtrations Let M be a compact smooth manifold with Morse functions f, g : M! R. Suppose kf gk 1 apple r. Writing [f ] t = f 1 ( 1, t], [g] t = g 1 ( 1, t], we have inclusions [f ] t [g] t+r, [g] t [f ] t+r. More generally, we have diagrams [f ] s / [f ] t / [f ] t+2r [f ] s+r / [f ] t+r [g] s+r / [g] t+r [g] s / [g] t / [g] t+2r which automatically commute, since all arrows are inclusion maps. Applying homology H,weobtainthemaps t, t and required relations for an r- interleaving between the persistence modules for the two sublevelset filtrations.
15 " < < = " Data! Persistence modules Sublevelset filtrations Let P be a compact polyhedron with piecewise linear maps f, g : P! R. Suppose kf gk 1 apple r. Writing [f ] t = f 1 ( 1, t], [g] t = g 1 ( 1, t], we have inclusions [f ] t [g] t+r, [g] t [f ] t+r. More generally, we have diagrams [f ] s / [f ] t / [f ] t+2r [f ] s+r / [f ] t+r [g] s+r / [g] t+r [g] s / [g] t / [g] t+2r which automatically commute, since all arrows are inclusion maps. Applying homology H,weobtainthemaps t, t and required relations for an r- interleaving between the persistence modules for the two sublevelset filtrations.
16 ! < < < " Data! Persistence modules Metric data (fixed set) Let X be a finite set and let d, e : X X! R be metrics. Suppose kd ek 1 apple r. Writing [d] t =VR((X, d), t) [e] t =VR((X, e), t) we have inclusions [d] t [e] t+r, [e] t [d] t+r. More generally, we have diagrams [d] s / [d] t / [d] t+2r [d] s+r / [d] t+r [e] s+r / [e] t+r [e] s / [e] t / [e] t+2r which automatically commute, since all arrows are inclusion maps. Applying homology H,weobtainthemaps t, t and required relations for an r-interleaving between the two Vietoris Rips homology persistence modules.
17 Data! Persistence modules More subtle The other two examples are slightly more subtle: Metric data (subsets of a fixed space) Let X, Y Z be finite subsets of a fixed metric space (Z, d). Metric data (general) Let (X, d), (Y, e) befinitemetricspaces.
18 Detour: Multivalued maps Multivalued maps Let X, Y be sets. A multivalued map F : X Y is a function F : X!P(Y) that carries each x 2 X to a nonempty subset of Y. Selections A selection from F is a function f : X! Y such that f (x) 2 F (x) for all x. Subordinate maps Let F, G : X Y.WesaythatG is subordinate to F,andwriteG F,if G(x) F (x) for all x. Equivalently, every selection from G is a selection from F.
19 Detour: Multivalued maps Multivalued simplicial maps Let S, T be simplicial complexes with vertex sets X, Y. A multivalued simplicial map F : S T is a multivalued map F : X Y, such that the image of every simplex 2 S F = S x2 F (x) (has the property that every finite subset) is a simplex of T. Induced homology map Let F : S T be a multivalued simplicial map. Then the maps f : H (S)! H (T ) induced by selections f are all equal to each other. Call this map F. Proof Indeed, any two selections are contiguous. Remark It is immediate that G F implies G = F.
20 Data! Persistence modules Metric data (subsets of a fixed space) Let X, Y Z be finite subsets of a fixed metric space (Z, d). Suppose d H(X, Y ) apple r/2. Thenwecandefinemultivaluedmaps: F : X Y ; G : Y X ; F (x) ={y 2 Y d(x, y) apple r/2}. G(y) ={x 2 Y d(y, x) apple r/2}. Using the triangle inequality, these define multivalued simplicial maps F :VR(X, t) VR(Y, t + r), G :VR(Y, t) VR(X, t + r). Applying homology, we get maps t = F : H (VR(X, t)) H (VR(Y, t + r)) t = G : H (VR(Y, t)) H (VR(X, t + r)) It remains to verify the required relations for an r-interleaving.
21 " " # # $ $ Data! Persistence modules Metric data (subsets of a fixed space) Let X, Y Z be finite subsets of a fixed metric space (Z, d). Write [X ] t =VR(X, t) and[y ] t =VR(Y, t). Then we have: [X ] s / [X ] t / [X ] t+2r [X ] s+r / : [X ] t+r : : : ; ; [Y ] s+r / [Y ] t+r [Y ] s / [Y ] t / [Y ] t+2r The two parallelograms commute. The triangles don t commute. However, the horizontal map in each triangle is subordinate to the composite of the two diagonal maps. Applying homology, the resulting diagrams do commute. Thus the maps t, t define an r-interleaving. Conclusion subset of a metric space / VR persistent homology is 2-Lipschitz.
22 Persistence Diagrams Diagrams A diagram in H is a multiset of points in the region of the extended plane Point metric H = {(p, q) 1< p < q apple +1} We use the `1 metric for points in this region: 8 >< max( p 1 p 2, q 1 q 2 ) if q 1, q 2 are finite d 1 ((p 1, q 1), (p 2, q 2)) = p 1 p 2 if q 1 = q 2 =+1 >: 1 otherwise We also note the distance from the diagonal: d 1 ((p, q), ) = 1 (q p) 2
23 Persistence Diagrams The bottleneck distance Let A, B be diagrams in H. An r-matching is a partial bijection between the multisets A, B such that: If a 2 A and b 2 B are matched, then d 1 (a, b) apple r. If a 2 A is not matched, then d 1 (a, ) apple r. If b 2 B is not matched, then d 1 (b, ) apple r. The bottleneck distance between A, B is d B(A, B) =inf r 0 r there exists an r-matching between A, B
24 Persistence modules! Persistence diagrams Algebraic Stability Theorem (Chazal et al. 2009) If two q-tame persistence modules are r-interleaved, then their persistence diagrams are r-matched. persistence module / persistence diagram is 1-Lipschitz Ingredients Box Lemma. Glisse Interpolation Lemma.
25 Persistence modules! Persistence diagrams Algebraic Stability Theorem (Chazal et al. 2009) If two q-tame persistence modules are r-interleaved, then their persistence diagrams are r-matched. Ingredients Box Lemma. persistence module / persistence diagram is 1-Lipschitz Glisse Interpolation Lemma. Box Lemma Let V, W be r-interleaved persistence modules with persistence diagrams A, B. Then the number of points of A inside any rectangle R is no greater than the number of points of B inside the rectangle R r obtained by thickening R on both sides by r (provided that R r does not meet the diagonal).
26 Persistence modules! Persistence diagrams Algebraic Stability Theorem (Chazal et al. 2009) If two q-tame persistence modules are r-interleaved, then their persistence diagrams are r-matched. Ingredients Box Lemma. persistence module / persistence diagram is 1-Lipschitz Glisse Interpolation Lemma. Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t.
27 > > / Persistence modules! Persistence diagrams Box Lemma Let V, W be r-interleaved persistence modules with persistence diagrams A, B. Then the number of points of A inside any rectangle R is no greater than the number of points of B inside the rectangle R r obtained by thickening R on both sides by r (provided that R r does not meet the diagonal). Outline Proof Let R =[a, b] [c, d] sothatr r =[a r, b + r] [c r, d + r]. Consider the diagram: a r vb a vc V a / b vd V b / c V c / V d b c r d W a r w a r b+r / W b+r / W c r w b+r c r w c r d+r W d+r
28 > > Persistence modules! Persistence diagrams Box Lemma Let V, W be r-interleaved persistence modules with persistence diagrams A, B. Then the number of points of A inside any rectangle R is no greater than the number of points of B inside the rectangle R r obtained by thickening R on both sides by r (provided that R r does not meet the diagonal). Outline Proof Let R =[a, b] [c, d] sothatr r =[a r, b + r] [c r, d + r]. Consider the diagram: V a v a b / V b V c v c d / V d a r b c r d W a r W b+r / W c r w b+r c r W d+r
29 Persistence modules! Persistence diagrams Box Lemma Let V, W be r-interleaved persistence modules with persistence diagrams A, B. Then the number of points of A inside any rectangle R is no greater than the number of points of B inside the rectangle R r obtained by thickening R on both sides by r (provided that R r does not meet the diagonal). Outline Proof Let R =[a, b] [c, d] sothatr r =[a r, b + r] [c r, d + r]. W a r / V a / V b / W b+r / W c r / V c / V d / W d+r Rectangle point counts are obtained by counting interval summands: # B \ R r : # A \ R :
30 Persistence modules! Persistence diagrams Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t. Remarks The lemma can be proved by direct algebraic construction; or by applying the Kan Extension Theorem from category theory. The lemma can be circumvented if the initial data can be interpolated.
31 Persistence modules! Persistence diagrams Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t. Remarks The lemma can be proved by direct algebraic construction; or by applying the Kan Extension Theorem from category theory. The lemma can be circumvented if the initial data can be interpolated. Simplex streams Let S be a simplicial complex and let f, g : S! R define simplex streams. Set f t =(1 t)f + tg.
32 Persistence modules! Persistence diagrams Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t. Remarks The lemma can be proved by direct algebraic construction; or by applying the Kan Extension Theorem from category theory. The lemma can be circumvented if the initial data can be interpolated. Sublevelset filtrations Let M be a compact smooth manifold with Morse functions f, g : M! R. Set f t =(1 t)f + tg.
33 Persistence modules! Persistence diagrams Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t. Remarks The lemma can be proved by direct algebraic construction; or by applying the Kan Extension Theorem from category theory. The lemma can be circumvented if the initial data can be interpolated. Sublevelset filtrations Let P be a compact polyhedron with piecewise linear maps f, g : P! R. Set f t =(1 t)f + tg.
34 Persistence modules! Persistence diagrams Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t. Remarks The lemma can be proved by direct algebraic construction; or by applying the Kan Extension Theorem from category theory. The lemma can be circumvented if the initial data can be interpolated. Metric data (fixed set) Let X be a finite set and let d, e : X X! R be metrics. Set d t =(1 t)d + te.
35 Persistence modules! Persistence diagrams Glisse Interpolation Lemma Let V, W be r-interleaved persistence modules. Then there is a 1-parameter family of persistence modules (V t t 2 [0, r]) such that V 0 = V and V r = W,withtheadditionalpropertythatV s, V t are s t -interleaved for all s, t. Remarks The lemma can be proved by direct algebraic construction; or by applying the Kan Extension Theorem from category theory. The lemma can be circumvented if the initial data can be interpolated. Metric data (subsets of a fixed space) Let X, Y Z be finite subsets of a fixed metric space (Z, d). Set X t = {(1 t)x + ty x 2 X, y 2 Y, d(x, y) apple d H(X, Y )}.
36 Persistence modules! Persistence diagrams Outline proof of Algebraic Stability Theorem The Box Lemma can be used to show that persistence module / persistence diagram is locally 1-Lipschitz (near persistence modules with a finite diagram). Local 1-Lipschitz implies global 1-Lipschitz, by applying a Heine Borel argument along a path provided by the Glisse Interpolation Lemma.
5. Persistence Diagrams
Vin de Silva Pomona College CBMS Lecture Series, Macalester College, June 207 Persistence modules Persistence modules A diagram of vector spaces and linear maps V 0 V... V n is called a persistence module.
More informationMetric Thickenings of Euclidean Submanifolds
Metric Thickenings of Euclidean Submanifolds Advisor: Dr. Henry Adams Committe: Dr. Chris Peterson, Dr. Daniel Cooley Joshua Mirth Masters Thesis Defense October 3, 2017 Introduction Motivation Can we
More informationSome Topics in Computational Topology. Yusu Wang. AMS Short Course 2014
Some Topics in Computational Topology Yusu Wang AMS Short Course 2014 Introduction Much recent developments in computational topology Both in theory and in their applications E.g, the theory of persistence
More informationSome Topics in Computational Topology
Some Topics in Computational Topology Yusu Wang Ohio State University AMS Short Course 2014 Introduction Much recent developments in computational topology Both in theory and in their applications E.g,
More informationInterleaving Distance between Merge Trees
Noname manuscript No. (will be inserted by the editor) Interleaving Distance between Merge Trees Dmitriy Morozov Kenes Beketayev Gunther H. Weber the date of receipt and acceptance should be inserted later
More informationHomology through Interleaving
Notes by Tamal K. Dey, OSU 91 Homology through Interleaving 32 Concept of interleaving A discrete set P R k is assumed to be a sample of a set X R k, if it lies near to it which we can quantify with the
More informationProximity of Persistence Modules and their Diagrams
Proximity of Persistence Modules and their Diagrams Frédéric Chazal David Cohen-Steiner Marc Glisse Leonidas J. Guibas Steve Y. Oudot November 28, 2008 Abstract Topological persistence has proven to be
More informationTopological Simplification in 3-Dimensions
Topological Simplification in 3-Dimensions David Letscher Saint Louis University Computational & Algorithmic Topology Sydney Letscher (SLU) Topological Simplification CATS 2017 1 / 33 Motivation Original
More informationPersistent Homology: Course Plan
Persistent Homology: Course Plan Andrey Blinov 16 October 2017 Abstract This is a list of potential topics for each class of the prospective course, with references to the literature. Each class will have
More informationPARAMETRIZED HOMOLOGY VIA ZIGZAG PERSISTENCE
PARAMETRIZED HOMOLOGY VIA ZIGZAG PERSISTENCE GUNNAR CARLSSON, VIN DE SILVA, SARA KALIŠNIK VEROVŠEK, DMITRIY MOROZOV Abstract. This paper develops the idea of homology for 1-parameter families of topological
More informationPersistence based Clustering
July 9, 2009 TGDA Workshop Persistence based Clustering Primoz Skraba joint work with Frédéric Chazal, Steve Y. Oudot, Leonidas J. Guibas 1-1 Clustering Input samples 2-1 Clustering Input samples Important
More informationarxiv: v2 [math.at] 10 Nov 2013
SKETCHES OF A PLATYPUS: A SURVEY OF PERSISTENT HOMOLOGY AND ITS ALGEBRAIC FOUNDATIONS MIKAEL VEJDEMO-JOHANSSON arxiv:1212.5398v2 [math.at] 10 Nov 2013 Computer Vision and Active Perception Laboratory KTH
More informationThe following slides heavily depend on [Chazal et al., 2016, Bubenik and Scott, 2014]
The following slides heavily depend on [Chazal et al., 2016, Bubenik and Scott, 2014] Plus a couple of examples from [Ghrist, 2014, Kleinberg, 2002, Carlsson and Mémoli, 2010] Section number are from [Chazal
More informationPersistent Local Systems
Persistent Local Systems Amit Patel Department of Mathematics Colorado State University joint work with Robert MacPherson School of Mathematics Institute for Advanced Study Abel Symposium on Topological
More informationMath 421, Homework #9 Solutions
Math 41, Homework #9 Solutions (1) (a) A set E R n is said to be path connected if for any pair of points x E and y E there exists a continuous function γ : [0, 1] R n satisfying γ(0) = x, γ(1) = y, and
More informationMetrics on Persistence Modules and an Application to Neuroscience
Metrics on Persistence Modules and an Application to Neuroscience Magnus Bakke Botnan NTNU November 28, 2014 Part 1: Basics The Study of Sublevel Sets A pair (M, f ) where M is a topological space and
More informationScalar Field Analysis over Point Cloud Data
DOI 10.1007/s00454-011-9360-x Scalar Field Analysis over Point Cloud Data Frédéric Chazal Leonidas J. Guibas Steve Y. Oudot Primoz Skraba Received: 19 July 2010 / Revised: 13 April 2011 / Accepted: 18
More informationarxiv: v4 [cs.cg] 2 Feb 2015
The Theory of the Interleaving Distance on Multidimensional Persistence Modules Michael Lesnick Institute for Mathematics and its Applications arxiv:1106.5305v4 [cs.cg] 2 Feb 2015 Abstract In 2009, Chazal
More informationPersistent homology and nonparametric regression
Cleveland State University March 10, 2009, BIRS: Data Analysis using Computational Topology and Geometric Statistics joint work with Gunnar Carlsson (Stanford), Moo Chung (Wisconsin Madison), Peter Kim
More informationPersistent Homology. 24 Notes by Tamal K. Dey, OSU
24 Notes by Tamal K. Dey, OSU Persistent Homology Suppose we have a noisy point set (data) sampled from a space, say a curve inr 2 as in Figure 12. Can we get the information that the sampled space had
More informationMetric shape comparison via multidimensional persistent homology
Metric shape comparison via multidimensional persistent homology Patrizio Frosini 1,2 1 Department of Mathematics, University of Bologna, Italy 2 ARCES - Vision Mathematics Group, University of Bologna,
More informationMultivariate Topological Data Analysis
Cleveland State University November 20, 2008, Duke University joint work with Gunnar Carlsson (Stanford), Peter Kim and Zhiming Luo (Guelph), and Moo Chung (Wisconsin Madison) Framework Ideal Truth (Parameter)
More informationLinear-Size Approximations to the Vietoris Rips Filtration
Linear-Size Approximations to the Vietoris Rips Filtration Donald R. Sheehy March 8, 2013 Abstract The Vietoris Rips filtration is a versatile tool in topological data analysis. It is a sequence of simplicial
More informationPersistent Homology of Data, Groups, Function Spaces, and Landscapes.
Persistent Homology of Data, Groups, Function Spaces, and Landscapes. Shmuel Weinberger Department of Mathematics University of Chicago William Benter Lecture City University of Hong Kong Liu Bie Ju Centre
More informationTHESIS METRIC THICKENINGS OF EUCLIDEAN SUBMANIFOLDS. Submitted by. Joshua R. Mirth. Department of Mathematics
THESIS METRIC THICKENINGS OF EUCLIDEAN SUBMANIFOLDS Submitted by Joshua R. Mirth Department of Mathematics In partial fulfillment of the requirements For the Degree of Master of Science Colorado State
More informationStability of Critical Points with Interval Persistence
Stability of Critical Points with Interval Persistence Tamal K. Dey Rephael Wenger Abstract Scalar functions defined on a topological space Ω are at the core of many applications such as shape matching,
More information24 PERSISTENT HOMOLOGY
24 PERSISTENT HOMOLOGY Herbert Edelsbrunner and Dmitriy Morozov INTRODUCTION Persistent homology was introduced in [ELZ00] and quickly developed into the most influential method in computational topology.
More informationTopological Data Analysis - II. Afra Zomorodian Department of Computer Science Dartmouth College
Topological Data Analysis - II Afra Zomorodian Department of Computer Science Dartmouth College September 4, 2007 1 Plan Yesterday: Motivation Topology Simplicial Complexes Invariants Homology Algebraic
More informationPersistent Homology. 128 VI Persistence
8 VI Persistence VI. Persistent Homology A main purpose of persistent homology is the measurement of the scale or resolution of a topological feature. There are two ingredients, one geometric, assigning
More informationGeometric Inference on Kernel Density Estimates
Geometric Inference on Kernel Density Estimates Bei Wang 1 Jeff M. Phillips 2 Yan Zheng 2 1 Scientific Computing and Imaging Institute, University of Utah 2 School of Computing, University of Utah Feb
More informationGraph Induced Complex: A Data Sparsifier for Homology Inference
Graph Induced Complex: A Data Sparsifier for Homology Inference Tamal K. Dey Fengtao Fan Yusu Wang Department of Computer Science and Engineering The Ohio State University October, 2013 Space, Sample and
More informationCOUNTEREXAMPLES TO THE COARSE BAUM-CONNES CONJECTURE. Nigel Higson. Unpublished Note, 1999
COUNTEREXAMPLES TO THE COARSE BAUM-CONNES CONJECTURE Nigel Higson Unpublished Note, 1999 1. Introduction Let X be a discrete, bounded geometry metric space. 1 Associated to X is a C -algebra C (X) which
More informationMaths 212: Homework Solutions
Maths 212: Homework Solutions 1. The definition of A ensures that x π for all x A, so π is an upper bound of A. To show it is the least upper bound, suppose x < π and consider two cases. If x < 1, then
More informationA Higher-Dimensional Homologically Persistent Skeleton
A Higher-Dimensional Homologically Persistent Skeleton Sara Kališnik Verovšek, Vitaliy Kurlin, Davorin Lešnik 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Abstract A data set is often given as a point cloud,
More informationarxiv: v1 [math.at] 27 Jul 2012
STATISTICAL TOPOLOGY USING PERSISTENCE LANDSCAPES PETER BUBENIK arxiv:1207.6437v1 [math.at] 27 Jul 2012 Abstract. We define a new descriptor for persistent homology, which we call the persistence landscape,
More informationAn Introduction to Topological Data Analysis
An Introduction to Topological Data Analysis Elizabeth Munch Duke University :: Dept of Mathematics Tuesday, June 11, 2013 Elizabeth Munch (Duke) CompTop-SUNYIT Tuesday, June 11, 2013 1 / 43 There is no
More informationRipser. Efficient Computation of Vietoris Rips Persistence Barcodes. Ulrich Bauer TUM
Ripser Efficient Computation of Vietoris Rips Persistence Barcodes Ulrich Bauer TUM March 23, 2017 Computational and Statistical Aspects of Topological Data Analysis Alan Turing Institute http://ulrich-bauer.org/ripser-talk.pdf
More informationSome Background Material
Chapter 1 Some Background Material In the first chapter, we present a quick review of elementary - but important - material as a way of dipping our toes in the water. This chapter also introduces important
More informationIntroduction to Algebraic and Geometric Topology Week 3
Introduction to Algebraic and Geometric Topology Week 3 Domingo Toledo University of Utah Fall 2017 Lipschitz Maps I Recall f :(X, d)! (X 0, d 0 ) is Lipschitz iff 9C > 0 such that d 0 (f (x), f (y)) apple
More informationChapter 4. Inverse Function Theorem. 4.1 The Inverse Function Theorem
Chapter 4 Inverse Function Theorem d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d d dd d d d d This chapter
More information7. Homotopy and the Fundamental Group
7. Homotopy and the Fundamental Group The group G will be called the fundamental group of the manifold V. J. Henri Poincaré, 895 The properties of a topological space that we have developed so far have
More informationLECTURE 15: COMPLETENESS AND CONVEXITY
LECTURE 15: COMPLETENESS AND CONVEXITY 1. The Hopf-Rinow Theorem Recall that a Riemannian manifold (M, g) is called geodesically complete if the maximal defining interval of any geodesic is R. On the other
More informationA metric approach to shape comparison via multidimensional persistence
A metric approach to shape comparison via multidimensional persistence Patrizio Frosini 1,2 1 Department of Mathematics, University of Bologna, Italy 2 ARCES - Vision Mathematics Group, University of Bologna,
More informationNonparametric regression for topology. applied to brain imaging data
, applied to brain imaging data Cleveland State University October 15, 2010 Motivation from Brain Imaging MRI Data Topology Statistics Application MRI Data Topology Statistics Application Cortical surface
More informationUsing the bottleneck distance to compare homological methods of detecting atmospheric rivers
Using the bottleneck distance to compare homological methods of detecting atmospheric rivers Kim Alexander Haugland January 17, 2018 1 Acknowledgements I wish to thank my supervisors Morten Brun and Bjørn
More informationCup product and intersection
Cup product and intersection Michael Hutchings March 28, 2005 Abstract This is a handout for my algebraic topology course. The goal is to explain a geometric interpretation of the cup product. Namely,
More informationEconomics 204 Summer/Fall 2011 Lecture 5 Friday July 29, 2011
Economics 204 Summer/Fall 2011 Lecture 5 Friday July 29, 2011 Section 2.6 (cont.) Properties of Real Functions Here we first study properties of functions from R to R, making use of the additional structure
More informationReview of Multi-Calculus (Study Guide for Spivak s CHAPTER ONE TO THREE)
Review of Multi-Calculus (Study Guide for Spivak s CHPTER ONE TO THREE) This material is for June 9 to 16 (Monday to Monday) Chapter I: Functions on R n Dot product and norm for vectors in R n : Let X
More informationAnalysis of Scalar Fields over Point Cloud Data
Analysis of Scalar Fields over Point Cloud Data Frédéric Chazal Leonidas J. Guibas Steve Y. Oudot Primoz Skraba Abstract Given a real-valued function f defined over some metric space X, is it possible
More informationBasic Properties of Metric and Normed Spaces
Basic Properties of Metric and Normed Spaces Computational and Metric Geometry Instructor: Yury Makarychev The second part of this course is about metric geometry. We will study metric spaces, low distortion
More informationStratification Learning through Homology Inference
Stratification Learning through Homology Inference Paul Bendich Sayan Mukherjee Institute of Science and Technology Austria Department of Statistical Science Klosterneuburg, Austria Duke University, Durham
More informationAnalysis of Scalar Fields over Point Cloud Data
INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET EN AUTOMATIQUE Analysis of Scalar Fields over Point Cloud Data Frédéric Chazal Leonidas J. Guibas Steve Y. Oudot Primoz Skraba N 6576 July 2008 Thème COM
More informationCanonical Stratifications along Bisheaves
Canonical Stratifications along Bisheaves Vidit Nanda and Amit Patel Abstract. A theory of bisheaves has been recently introduced to measure the homological stability of fibers of maps to manifolds. A
More informationEXPLICIT l 1 -EFFICIENT CYCLES AND AMENABLE NORMAL SUBGROUPS
EXPLICIT l 1 -EFFICIENT CYCLES AND AMENABLE NORMAL SUBGROUPS CLARA LÖH Abstract. By Gromov s mapping theorem for bounded cohomology, the projection of a group to the quotient by an amenable normal subgroup
More informationReal Analysis Notes. Thomas Goller
Real Analysis Notes Thomas Goller September 4, 2011 Contents 1 Abstract Measure Spaces 2 1.1 Basic Definitions........................... 2 1.2 Measurable Functions........................ 2 1.3 Integration..............................
More informationPERVERSE SHEAVES ON A TRIANGULATED SPACE
PERVERSE SHEAVES ON A TRIANGULATED SPACE A. POLISHCHUK The goal of this note is to prove that the category of perverse sheaves constructible with respect to a triangulation is Koszul (i.e. equivalent to
More information2 (Bonus). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure?
MA 645-4A (Real Analysis), Dr. Chernov Homework assignment 1 (Due 9/5). Prove that every countable set A is measurable and µ(a) = 0. 2 (Bonus). Let A consist of points (x, y) such that either x or y is
More informationLARGE AND SMALL GROUP HOMOLOGY
LARGE AND SMALL GROUP HOMOLOGY MICHAEL BRUNNBAUER AND BERNHARD HANKE ABSTRACT. For several instances of metric largeness like enlargeability or having hyperspherical universal covers, we construct non-large
More informationTopological complexity of motion planning algorithms
Topological complexity of motion planning algorithms Mark Grant mark.grant@ed.ac.uk School of Mathematics - University of Edinburgh 31st March 2011 Mark Grant (University of Edinburgh) TC of MPAs 31st
More informationReeb Graphs: Approximation and Persistence
Reeb Graphs: Approximation and Persistence Tamal K. Dey Yusu Wang Abstract Given a continuous function f : X IR on a topological space X, its level set f 1 (a) changes continuously as the real value a
More informationCW-complexes. Stephen A. Mitchell. November 1997
CW-complexes Stephen A. Mitchell November 1997 A CW-complex is first of all a Hausdorff space X equipped with a collection of characteristic maps φ n α : D n X. Here n ranges over the nonnegative integers,
More informationarxiv: v4 [math.at] 23 Jan 2015
STATISTICAL TOPOLOGICAL DATA ANALYSIS USING PERSISTENCE LANDSCAPES PETER BUBENIK arxiv:1207.6437v4 [math.at] 23 Jan 2015 Abstract. We define a new topological summary for data that we call the persistence
More informationProblem Set 2: Solutions Math 201A: Fall 2016
Problem Set 2: s Math 201A: Fall 2016 Problem 1. (a) Prove that a closed subset of a complete metric space is complete. (b) Prove that a closed subset of a compact metric space is compact. (c) Prove that
More informationAnalysis of Scalar Fields over Point Cloud Data
Analysis of Scalar Fields over Point Cloud Data Frédéric Chazal, Leonidas J. Guibas, Steve Y. Oudot, Primoz Skraba To cite this version: Frédéric Chazal, Leonidas J. Guibas, Steve Y. Oudot, Primoz Skraba.
More informationLecture 5 - Hausdorff and Gromov-Hausdorff Distance
Lecture 5 - Hausdorff and Gromov-Hausdorff Distance August 1, 2011 1 Definition and Basic Properties Given a metric space X, the set of closed sets of X supports a metric, the Hausdorff metric. If A is
More informationIntroduction to Topology
Introduction to Topology Randall R. Holmes Auburn University Typeset by AMS-TEX Chapter 1. Metric Spaces 1. Definition and Examples. As the course progresses we will need to review some basic notions about
More informationA Higher-Dimensional Homologically Persistent Skeleton
arxiv:1701.08395v5 [math.at] 10 Oct 2017 A Higher-Dimensional Homologically Persistent Skeleton Sara Kališnik Verovšek, Vitaliy Kurlin, Davorin Lešnik Abstract Real data is often given as a point cloud,
More informationLecture 1. Toric Varieties: Basics
Lecture 1. Toric Varieties: Basics Taras Panov Lomonosov Moscow State University Summer School Current Developments in Geometry Novosibirsk, 27 August1 September 2018 Taras Panov (Moscow University) Lecture
More informationTHE INVERSE FUNCTION THEOREM
THE INVERSE FUNCTION THEOREM W. PATRICK HOOPER The implicit function theorem is the following result: Theorem 1. Let f be a C 1 function from a neighborhood of a point a R n into R n. Suppose A = Df(a)
More informationg 2 (x) (1/3)M 1 = (1/3)(2/3)M.
COMPACTNESS If C R n is closed and bounded, then by B-W it is sequentially compact: any sequence of points in C has a subsequence converging to a point in C Conversely, any sequentially compact C R n is
More informationBirth and death in discrete Morse theory
Birth and death in discrete Morse theory Neža Mramor Kosta joint work with Henry C. King, Kevin Knudson University of Ljubljana, Slovenia ACAT Bremen 2013 Motivation Discrete Morse theory Parametric discrete
More informationarxiv: v2 [math.mg] 25 Oct 2017
The Hausdorff Mapping Is Nonexpanding arxiv:1710.08472v2 [math.mg] 25 Oct 2017 Ivan A. Mikhaylov Abstract In the present paper we investigate the properties of the Hausdorff mapping H, which takes each
More informationLECTURE Itineraries We start with a simple example of a dynamical system obtained by iterating the quadratic polynomial
LECTURE. Itineraries We start with a simple example of a dynamical system obtained by iterating the quadratic polynomial f λ : R R x λx( x), where λ [, 4). Starting with the critical point x 0 := /2, we
More informationMultiscale Mapper: Topological Summarization via Codomain Covers
Multiscale Mapper: Topological Summarization via Codomain Covers Tamal K. Dey, Facundo Mémoli, Yusu Wang Abstract Summarizing topological information from datasets and maps defined on them is a central
More informationStanford Mathematics Department Math 205A Lecture Supplement #4 Borel Regular & Radon Measures
2 1 Borel Regular Measures We now state and prove an important regularity property of Borel regular outer measures: Stanford Mathematics Department Math 205A Lecture Supplement #4 Borel Regular & Radon
More informationMATH 51H Section 4. October 16, Recall what it means for a function between metric spaces to be continuous:
MATH 51H Section 4 October 16, 2015 1 Continuity Recall what it means for a function between metric spaces to be continuous: Definition. Let (X, d X ), (Y, d Y ) be metric spaces. A function f : X Y is
More informationarxiv: v5 [math.mg] 6 Sep 2018
PERSISTENT HOMOLOGY AND THE UPPER BOX DIMENSION BENJAMIN SCHWEINHART arxiv:1802.00533v5 [math.mg] 6 Sep 2018 Abstract. We introduce a fractal dimension for a metric space based on the persistent homology
More information1 k x k. d(x, y) =sup k. y k = max
1 Lecture 13: October 8 Urysohn s metrization theorem. Today, I want to explain some applications of Urysohn s lemma. The first one has to do with the problem of characterizing metric spaces among all
More informationLecture 4: Completion of a Metric Space
15 Lecture 4: Completion of a Metric Space Closure vs. Completeness. Recall the statement of Lemma??(b): A subspace M of a metric space X is closed if and only if every convergent sequence {x n } X satisfying
More informationThe fundamental group of a visual boundary versus the fundamental group at infinity
The fundamental group of a visual boundary versus the fundamental group at infinity Greg Conner and Hanspeter Fischer June 2001 There is a natural homomorphism from the fundamental group of the boundary
More informationAn introduction to discrete Morse theory
An introduction to discrete Morse theory Henry Adams December 11, 2013 Abstract This talk will be an introduction to discrete Morse theory. Whereas standard Morse theory studies smooth functions on a differentiable
More informationA new proof of Gromov s theorem on groups of polynomial growth
A new proof of Gromov s theorem on groups of polynomial growth Bruce Kleiner Courant Institute NYU Groups as geometric objects Let G a group with a finite generating set S G. Assume that S is symmetric:
More informationPatrizio Frosini. Department of Mathematics and ARCES, University of Bologna
An observer-oriented approach to topological data analysis Part 2: The algebra of group invariant non-expansive operators and its application in the project GIPHOD - Group Invariant Persistent Homology
More informationGeometric Inference for Probability distributions
Geometric Inference for Probability distributions F. Chazal 1 D. Cohen-Steiner 2 Q. Mérigot 2 1 Geometrica, INRIA Saclay, 2 Geometrica, INRIA Sophia-Antipolis 2009 July 8 F. Chazal, D. Cohen-Steiner, Q.
More informationEquivalence of the Combinatorial Definition (Lecture 11)
Equivalence of the Combinatorial Definition (Lecture 11) September 26, 2014 Our goal in this lecture is to complete the proof of our first main theorem by proving the following: Theorem 1. The map of simplicial
More informationThe Persistent Homology of Distance Functions under Random Projection
The Persistent Homology of Distance Functions under Random Projection Donald R. Sheehy University of Connecticut don.r.sheehy@gmail.com Abstract Given n points P in a Euclidean space, the Johnson-Linden-strauss
More informationDIFFERENTIAL GEOMETRY, LECTURE 16-17, JULY 14-17
DIFFERENTIAL GEOMETRY, LECTURE 16-17, JULY 14-17 6. Geodesics A parametrized line γ : [a, b] R n in R n is straight (and the parametrization is uniform) if the vector γ (t) does not depend on t. Thus,
More informationComparing persistence diagrams through complex vectors
Comparing persistence diagrams through complex vectors Barbara Di Fabio ARCES, University of Bologna Final Project Meeting Vienna, July 9, 2015 B. Di Fabio, M. Ferri, Comparing persistence diagrams through
More informationHalf of Final Exam Name: Practice Problems October 28, 2014
Math 54. Treibergs Half of Final Exam Name: Practice Problems October 28, 24 Half of the final will be over material since the last midterm exam, such as the practice problems given here. The other half
More informationOverview of normed linear spaces
20 Chapter 2 Overview of normed linear spaces Starting from this chapter, we begin examining linear spaces with at least one extra structure (topology or geometry). We assume linearity; this is a natural
More informationDefinitions and Properties of R N
Definitions and Properties of R N R N as a set As a set R n is simply the set of all ordered n-tuples (x 1,, x N ), called vectors. We usually denote the vector (x 1,, x N ), (y 1,, y N ), by x, y, or
More informationSOME REMARKS ON THE TOPOLOGY OF HYPERBOLIC ACTIONS OF R n ON n-manifolds
SOME REMARKS ON THE TOPOLOGY OF HYPERBOLIC ACTIONS OF R n ON n-manifolds DAMIEN BOULOC Abstract. This paper contains some more results on the topology of a nondegenerate action of R n on a compact connected
More information3 (Due ). Let A X consist of points (x, y) such that either x or y is a rational number. Is A measurable? What is its Lebesgue measure?
MA 645-4A (Real Analysis), Dr. Chernov Homework assignment 1 (Due ). Show that the open disk x 2 + y 2 < 1 is a countable union of planar elementary sets. Show that the closed disk x 2 + y 2 1 is a countable
More informationTopological properties
CHAPTER 4 Topological properties 1. Connectedness Definitions and examples Basic properties Connected components Connected versus path connected, again 2. Compactness Definition and first examples Topological
More informationarxiv: v2 [cs.cg] 15 Feb 2013
MATHEMATICS OF COMPUTATION Volume 00, Number 0, Pages 000 000 S 0025-5718(XX)0000-0 PERSISTENCE MODULES: ALGEBRA AND ALGORITHMS PRIMOZ SKRABA Jožef Stefan Institute, Ljubljana, Slovenia arxiv:1302.2015v2
More informationOptimal Reconstruction Might be Hard
Optimal Reconstruction Might be Hard Dominique Attali, André Lieutier To cite this version: Dominique Attali, André Lieutier. Optimal Reconstruction Might be Hard. Discrete and Computational Geometry,
More informationPersistence modules in symplectic topology
Leonid Polterovich, Tel Aviv Cologne, 2017 based on joint works with Egor Shelukhin, Vukašin Stojisavljević and a survey (in progress) with Jun Zhang Morse homology f -Morse function, ρ-generic metric,
More informationSeries. TDA Learning Seminar. Koundinya Vajjha. June 1, Topological Data Analysis of Financial Time. Series.
of of TDA Learning Seminar June 1, 2018 References of M. Gidea,Y.Katz. Topological data analysis of financial time series: of crashes. Physica A: Statistical Mechanics and its Applications, 491:820-834,
More informationReal Analysis Chapter 4 Solutions Jonathan Conder
2. Let x, y X and suppose that x y. Then {x} c is open in the cofinite topology and contains y but not x. The cofinite topology on X is therefore T 1. Since X is infinite it contains two distinct points
More informationGeodesic Delaunay Triangulations in Bounded Planar Domains
Geodesic Delaunay Triangulations in Bounded Planar Domains Leonidas J. Guibas Dept. of Computer Science Stanford University Stanford, CA guibas@cs.stanford.edu Jie Gao Dept. of Computer Science Stony Brook
More information