Graph Induced Complex: A Data Sparsifier for Homology Inference
|
|
- Darlene Newton
- 5 years ago
- Views:
Transcription
1 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
2 Space, Sample and Complex Space (IMA) Graph Induced Complex October / 25
3 Space, Sample and Complex Space Samples (IMA) Graph Induced Complex October / 25
4 Space, Sample and Complex Space Samples Rips Complex (IMA) Graph Induced Complex October / 25
5 Complexes Delaunay complex difficult to compute in high dimensional spaces; (IMA) Graph Induced Complex October / 25
6 Complexes Delaunay complex difficult to compute in high dimensional spaces; Vietoris-Rips complex too large (5000 points in R 3 millions of simplicies); (IMA) Graph Induced Complex October / 25
7 Complexes Delaunay complex difficult to compute in high dimensional spaces; Vietoris-Rips complex too large (5000 points in R 3 millions of simplicies); Čech complex difficult to compute; also large; (IMA) Graph Induced Complex October / 25
8 Complexes Delaunay complex difficult to compute in high dimensional spaces; Vietoris-Rips complex too large (5000 points in R 3 millions of simplicies); Čech complex difficult to compute; also large; Witness complex manageable size; lack topological inference; (IMA) Graph Induced Complex October / 25
9 Subsamples (IMA) Graph Induced Complex October / 25
10 Subsamples Q (IMA) Graph Induced Complex October / 25
11 Subsamples R r (Q) (IMA) Graph Induced Complex October / 25
12 Subsamples R r (Q) Witness Complex (IMA) Graph Induced Complex October / 25
13 Subsamples R r (Q) Witness Complex (IMA) Graph Induced Complex October / 25
14 Our solution R r (P ) (IMA) Graph Induced Complex October / 25
15 Our solution R r (P ) G r (P, Q) (IMA) Graph Induced Complex October / 25
16 Input Assumptions P finite point set; (P, d) metric space; G(P ) be a graph; (IMA) Graph Induced Complex October / 25
17 Subsampling Q P a subset; ν(p) : the closest point of p P in Q; (IMA) Graph Induced Complex October / 25
18 Graph Induced Complex Graph induced complex G(P, Q, d) : {q 1,..., q k+1 } Q; a (k + 1)-clique in G(P ) with vertices p1,..., p k+1 ; ν(pi ) = q i ; (IMA) Graph Induced Complex October / 25
19 Graph Induced Complex Graph induced complex G(P, Q, d) : {q 1,..., q k+1 } Q; a (k + 1)-clique in G(P ) with vertices p1,..., p k+1 ; ν(pi ) = q i ; Remark: G(P, Q, d) depends on the metric d; Euclidean distance d E ; Graph based distance d G ; (IMA) Graph Induced Complex October / 25
20 Subsample Q of P δ-sample of P p P, q Q such that d(p, q) δ; (IMA) Graph Induced Complex October / 25
21 Subsample Q of P δ-sample of P p P, q Q such that d(p, q) δ; δ-sparse d(p, q) δ for any two distinct points p, q Q; (IMA) Graph Induced Complex October / 25
22 Subsample Q of P δ-sample of P p P, q Q such that d(p, q) δ; δ-sparse d(p, q) δ for any two distinct points p, q Q; Computing δ-sparse δ-sample Q iterative procedure; (IMA) Graph Induced Complex October / 25
23 Subsample Q of P δ-sample of P p P, q Q such that d(p, q) δ; δ-sparse d(p, q) δ for any two distinct points p, q Q; Computing δ-sparse δ-sample Q iterative procedure; δ Q i (IMA) Graph Induced Complex October / 25
24 Subsample Q of P δ-sample of P p P, q Q such that d(p, q) δ; δ-sparse d(p, q) δ for any two distinct points p, q Q; Computing δ-sparse δ-sample Q iterative procedure; δ δ q i+1 Q i Q i+1 (IMA) Graph Induced Complex October / 25
25 Subsample Q of P δ-sample of P p P, q Q such that d(p, q) δ; δ-sparse d(p, q) δ for any two distinct points p, q Q; Computing δ-sparse δ-sample Q iterative procedure; δ δ δ q i+1 Q i Q i+1 Q i+1 (IMA) Graph Induced Complex October / 25
26 Results H 1 inference in R n by d E and d G ; (IMA) Graph Induced Complex October / 25
27 Results H 1 inference in R n by d E and d G ; Surface reconstruction in R 3 ; (IMA) Graph Induced Complex October / 25
28 Results H 1 inference in R n by d E and d G ; Surface reconstruction in R 3 ; Improved H 1 inference in R n by d G from a lean subsample ; Complex size in log scale 14 Witness complex Rips complex 12 Graph induced complex Klein bottle in R 4 Dimension of H 1 14 Witness complex 12 Rips complex Graph induced complex δ δ (IMA) Graph Induced Complex October / 25
29 Results H 1 inference in R n by d E and d G ; Surface reconstruction in R 3 ; Improved H 1 inference in R n by d G from a lean subsample ; Complex size in log scale 14 Witness complex Rips complex 12 Graph induced complex Klein bottle in R 4 Dimension of H 1 14 Witness complex 12 Rips complex Graph induced complex δ δ Topological inference for compact sets in R n ; G α (P, Q, d) G 4(α+2δ) (P, Q, d) (IMA) Graph Induced Complex October / 25
30 Results H 1 inference in R n by d E and d G ; Surface reconstruction in R 3 ; Improved H 1 inference in R n by d G from a lean subsample ; Complex size in log scale 14 Witness complex Rips complex 12 Graph induced complex Klein bottle in R 4 Dimension of H 1 14 Witness complex 12 Rips complex Graph induced complex δ δ Topological inference for compact sets in R n ; G α (P, Q, d) G 4(α+2δ) (P, Q, d) (IMA) Graph Induced Complex October / 25
31 Simplicial map f : K L simplicial map for every simplex σ = {v 1, v 2,..., v k } K, f(σ) = {f(v 1 ), f(v 2 ),..., f(v k )} is a simplex in L (IMA) Graph Induced Complex October / 25
32 Simplicial map f : K L simplicial map for every simplex σ = {v 1, v 2,..., v k } K, f(σ) = {f(v 1 ), f(v 2 ),..., f(v k )} is a simplex in L w w u K v x f u L y x (IMA) Graph Induced Complex October / 25
33 Simplicial map f : K L simplicial map for every simplex σ = {v 1, v 2,..., v k } K, f(σ) = {f(v 1 ), f(v 2 ),..., f(v k )} is a simplex in L w w u K v x f u L y x [u] [u] [v] [u] [w] [w] [x] [x] [uv] [u] [uw] [uw] [vw] [uw] [vx] [vx] [uvw] [uw] (IMA) Graph Induced Complex October / 25
34 Contiguous maps f, g : K 1 K 2 two simplicial maps are contiguous for any simplex σ K 1, the simplices f(σ) and g(σ) are faces of a common simplex in K 2. (IMA) Graph Induced Complex October / 25
35 Contiguous maps f, g : K 1 K 2 two simplicial maps are contiguous for any simplex σ K 1, the simplices f(σ) and g(σ) are faces of a common simplex in K 2. u v x f g y z f([v]) = [y] g([v]) = [z], f([uv]) = [xy] g([uv]) = [xz] (IMA) Graph Induced Complex October / 25
36 Contiguous maps f, g : K 1 K 2 two simplicial maps are contiguous for any simplex σ K 1, the simplices f(σ) and g(σ) are faces of a common simplex in K 2. u v x f g y z f([v]) = [y] g([v]) = [z], f([uv]) = [xy] g([uv]) = [xz] Fact If f : K 1 K 2 and g : K 1 K 2 are contiguous, then the induced homomorphisms f : H n(k 1 ) H n(k 2 ) and g : H n(k 1 ) H n(k 2 ) are equal. (IMA) Graph Induced Complex October / 25
37 H 1 inference in R n Sample P from a space M; (IMA) Graph Induced Complex October / 25
38 H 1 inference in R n Sample P from a space M; Subsample Q : δ-sample, δ-sparse; (IMA) Graph Induced Complex October / 25
39 H 1 inference in R n Sample P from a space M; Subsample Q : δ-sample, δ-sparse; G α (P ) = 1-skeleton of R α (P ) (IMA) Graph Induced Complex October / 25
40 H 1 inference in R n Sample P from a space M; Subsample Q : δ-sample, δ-sparse; G α (P ) = 1-skeleton of R α (P ) The GIC G α (P, Q) Built on G α (P ) (IMA) Graph Induced Complex October / 25
41 H 1 inference in R n Sample P from a space M; Subsample Q : δ-sample, δ-sparse; G α (P ) = 1-skeleton of R α (P ) The GIC G α (P, Q) Built on G α (P ) h : R α (P ) G α (P, Q) simplicial map (IMA) Graph Induced Complex October / 25
42 H 1 inference in R n Sample P from a space M; Subsample Q : δ-sample, δ-sparse; G α (P ) = 1-skeleton of R α (P ) The GIC G α (P, Q) Built on G α (P ) h : R α (P ) G α (P, Q) simplicial map h : H(R α (P )) H(G α (P, Q)) isomorphism? (IMA) Graph Induced Complex October / 25
43 Contiguous maps between Rips and GIC R α (P ) h G α (P, Q, d) (IMA) Graph Induced Complex October / 25
44 Contiguous maps between Rips and GIC R α (P ) h R α+2δ (P ) G α (P, Q, d) (IMA) Graph Induced Complex October / 25
45 Contiguous maps between Rips and GIC i R α (P ) h R α+2δ (P ) G α (P, Q, d) (IMA) Graph Induced Complex October / 25
46 Contiguous maps between Rips and GIC R α (P ) h j R α+2δ (P ) G α (P, Q, d) (IMA) Graph Induced Complex October / 25
47 Contiguous maps between Rips and GIC i R α (P ) h j R α+2δ (P ) G α (P, Q, d) (IMA) Graph Induced Complex October / 25
48 Contiguous maps between Rips and GIC i R α (P ) h j R α+2δ (P ) G α (P, Q, d) j h contiguous to i; (IMA) Graph Induced Complex October / 25
49 Contiguous maps between Rips and GIC i R α (P ) h j R α+2δ (P ) j h contiguous to i; (j h) = i ; G α (P, Q, d) (j h), i : H(R α (P )) H(R α+2δ (P )) (IMA) Graph Induced Complex October / 25
50 Isomorphism of h Contiguous maps : j h = i ; (IMA) Graph Induced Complex October / 25
51 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; (IMA) Graph Induced Complex October / 25
52 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; h is injective; (IMA) Graph Induced Complex October / 25
53 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; h is injective; When h surjective? (IMA) Graph Induced Complex October / 25
54 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; h is injective; When h surjective? Higher dimensional homology (dim > 1), UNKNOWN; (IMA) Graph Induced Complex October / 25
55 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; h is injective; When h surjective? Higher dimensional homology (dim > 1), UNKNOWN; Positive answers for the 1-st dimensional homology : (IMA) Graph Induced Complex October / 25
56 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; h is injective; When h surjective? Higher dimensional homology (dim > 1), UNKNOWN; Positive answers for the 1-st dimensional homology : Euclidean distance de in R 3 ; (IMA) Graph Induced Complex October / 25
57 Isomorphism of h Contiguous maps : j h = i ; i isomorphism; i.e. P sampled from a manifold with positive reach; Prop 4.1 of [Dey and Wang 11] i : H 1 (R α (P )) H 1 (R β (P )) isomorphism; h is injective; When h surjective? Higher dimensional homology (dim > 1), UNKNOWN; Positive answers for the 1-st dimensional homology : Euclidean distance de in R 3 ; Graph distance dg in R n ; h : H 1 (R α (P )) H 1 (G α (P, Q)) isomorphism (IMA) Graph Induced Complex October / 25
58 Isomorphism of h Theorem P : an ɛ-sample of a surface with positive reach ρ in R 3 ; Q: a δ-sparse δ-sample of (P, d E ); ɛ ρ, 12ɛ α ρ, and 8ɛ δ ρ; h : H 1 (R α (P )) H 1 (G α (P, Q, d E )) isomorphism. (IMA) Graph Induced Complex October / 25
59 Isomorphism of h Theorem P : an ɛ-sample of a surface with positive reach ρ in R 3 ; Q: a δ-sparse δ-sample of (P, d E ); ɛ ρ, 12ɛ α ρ, and 8ɛ δ ρ; h : H 1 (R α (P )) H 1 (G α (P, Q, d E )) isomorphism. Theorem P : an ɛ-sample of manifold M with positive reach ρ; Q: a δ-sample of (P, d G ); 4ɛ α, δ 1 3 ρ, 3 5 h : H 1 (R α (P )) H 1 (G α (P, Q, d G )) isomorphism. (IMA) Graph Induced Complex October / 25
60 Improved H 1 Inference Homological loop feature size for simplicial complex K hlfs(k) = 1 inf{ c, c is non null-homologous 1-cycle in K}. 2 (IMA) Graph Induced Complex October / 25
61 Improved H 1 Inference Homological loop feature size for simplicial complex K hlfs(k) = 1 inf{ c, c is non null-homologous 1-cycle in K}. 2 ρ hlfs (IMA) Graph Induced Complex October / 25
62 Improved H 1 Inference Homological loop feature size for simplicial complex K hlfs(k) = 1 inf{ c, c is non null-homologous 1-cycle in K}. 2 ρ hlfs Theorem If Q is a δ-sample of (P, d G ) for δ < 1 2 hlfs(rα (P )) 1 α, then 2 h : H 1 (R α (P )) H 1 (G α (P, Q, d G )) is an isomorphism. (IMA) Graph Induced Complex October / 25
63 Experimental Result for Improved H 1 Inference Dimension of H Witness complex Rips complex Graph induced complex Complex size in log scale Witness complex Rips complex Graph induced complex δ (a) δ (b) Klein bottle in R 4 with P = 40, 000; GIC size : 154 (δ = 1.0) (IMA) Graph Induced Complex October / 25
64 Surface Reconstruction Crust [AB99] and Cocone [ACDL00]: Compute a subcomplex T Del P ; Argue T contains the restricted Delaunay triangulation Del M P ; Prune T to output a 2-manifold; (IMA) Graph Induced Complex October / 25
65 Surface Reconstruction by GIC in R 3 Restricted Delaunay triangulation Del M Q G(P, Q, d E ); (IMA) Graph Induced Complex October / 25
66 Surface Reconstruction by GIC in R 3 Restricted Delaunay triangulation Del M Q G(P, Q, d E ); Problem: GIC not embedded in R 3 ; t 2 t 1 (IMA) Graph Induced Complex October / 25
67 Surface Reconstruction by GIC in R 3 Restricted Delaunay triangulation Del M Q G(P, Q, d E ); Problem: GIC not embedded in R 3 ; t 2 t 1 Cleaning If V is the vertex set of t 1 and t 2 together, then at least one of t 1 and t 2 is not in Del V. The triangle which is not in Del V cannot be in Del Q as well. (IMA) Graph Induced Complex October / 25
68 Theorem P : ε-sample Q: δ-sparse, δ-sample of P 8ε δ 2 ρ, α 8ε 27 A triangulation T G α (P, Q, d E ) can be computed. FERTILITY BOTIJO mesh GIC mesh GIC 0-dim dim dim dim P = 1, 575, 055 for FERTILITY; P = 1, 049, 892 for BOTIJO; (IMA) Graph Induced Complex October / 25
69 GIC for Compact Sets Go beyond manifold and H 1 ; (IMA) Graph Induced Complex October / 25
70 GIC for Compact Sets Go beyond manifold and H 1 ; X R n compact set, and X λ offset; (IMA) Graph Induced Complex October / 25
71 GIC for Compact Sets Go beyond manifold and H 1 ; X R n compact set, and X λ offset; Homology of X λ captured by persistence of Rips complex on its samples P ; (IMA) Graph Induced Complex October / 25
72 GIC for Compact Sets Go beyond manifold and H 1 ; X R n compact set, and X λ offset; Homology of X λ captured by persistence of Rips complex on its samples P ; Interleave GIC with Rips complexes : persistence of GIC gives homology of X λ ; (IMA) Graph Induced Complex October / 25
73 GIC for Compact Sets Go beyond manifold and H 1 ; X R n compact set, and X λ offset; Homology of X λ captured by persistence of Rips complex on its samples P ; Interleave GIC with Rips complexes : persistence of GIC gives homology of X λ ; R α (P ) i 1 R h 1 α+2δ (P ) i 2 R 4β (P ) i 3 R 4β+2δ (P ) j 2 j 1 h 2 G α (P, Q, d) h G 4β (P, Q, d) Q δ-sparse δ-sample of P ; Q δ -sparse δ -sample of P with δ > δ; Denote β = α + 2δ; (IMA) Graph Induced Complex October / 25
74 GIC for Compact Sets Above diagram gives sequence H k (R α (P )) h 1 H k (G α (P, Q, d)) j 1 H k (R α+2δ (P )) i 2 H k (R 4β (P )) h 2 H k (G 4β (P, Q, d)) j 2 H k (R 4β+2δ (P )) (IMA) Graph Induced Complex October / 25
75 GIC for Compact Sets Above diagram gives sequence H k (R α (P )) h 1 H k (G α (P, Q, d)) j 1 H k (R α+2δ (P )) i 2 H k (R 4β (P )) h 2 H k (G 4β (P, Q, d)) j 2 H k (R 4β+2δ (P )) h = (h 2 i 2 j 1 ) simplicial map; Algorithms for persistence of simplicial maps [DFW]; (IMA) Graph Induced Complex October / 25
76 GIC for Compact Sets Above diagram gives sequence H k (R α (P )) h 1 H k (G α (P, Q, d)) j 1 H k (R α+2δ (P )) i 2 H k (R 4β (P )) h 2 H k (G 4β (P, Q, d)) j 2 H k (R 4β+2δ (P )) h = (h 2 i 2 j 1 ) simplicial map; Algorithms for persistence of simplicial maps [DFW]; Theorem P : an ɛ-sample of a compact set (X, d E ); Q: a δ-sparse δ-sample of (P, d) (d = d E or d G ); Q : a δ -sparse δ -sample of (P, d) (δ > δ); 0 < ɛ < 1 9 wfs(x), 2ɛ α 1 4 (wfs(x) ɛ) and (α + 2δ) δ 1 (wfs(x) ɛ), 4 im h = Hk (X λ ) (0 < λ < wfs(x)) (IMA) Graph Induced Complex October / 25
77 Conclusion Software for constructing GIC Available at authors webpages; (IMA) Graph Induced Complex October / 25
78 Conclusion Software for constructing GIC Available at authors webpages; Future work Go beyond of H 1 for h ; h : H n (R α (P )) H n (G α (P, Q)) (n 2) isomorphism? (IMA) Graph Induced Complex October / 25
79 Conclusion Software for constructing GIC Available at authors webpages; Future work Go beyond of H 1 for h ; h : H n (R α (P )) H n (G α (P, Q)) (n 2) isomorphism? Potential use in topological data analysis; (IMA) Graph Induced Complex October / 25
80 (IMA) Graph Induced Complex October / 25
Some 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 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 informationSimBa: An Efficient Tool for Approximating Rips-filtration Persistence via Simplicial Batch-collapse
SimBa: An Efficient Tool for Approximating Rips-filtration Persistence via Simplicial Batch-collapse Tamal K. Dey Department of Computer Science and Engineering The Ohio State University 2016 Joint work
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 information7.3 Singular Homology Groups
184 CHAPTER 7. HOMOLOGY THEORY 7.3 Singular Homology Groups 7.3.1 Cycles, Boundaries and Homology Groups We can define the singular p-chains with coefficients in a field K. Furthermore, we can define the
More informationDimension Detection by Local Homology
Dimension Detection by Local Homology Tamal K. Dey Fengtao Fan Yusu Wang Abstract Detecting the dimension of a hidden manifold from a point sample has become an important problem in the current data-driven
More informationParameter-free Topology Inference and Sparsification for Data on Manifolds
Parameter-free Topology Inference and Sparsification for Data on Manifolds Tamal K. Dey Zhe Dong Yusu Wang Abstract In topology inference from data, current approaches face two major problems. One concerns
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 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 informationPart II. Algebraic Topology. Year
Part II Year 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2017 Paper 3, Section II 18I The n-torus is the product of n circles: 5 T n = } S 1. {{.. S } 1. n times For all n 1 and 0
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 informationApproximating Cycles in a Shortest Basis of the First Homology Group from Point Data
Approximating Cycles in a Shortest Basis of the First Homology Group from Point Data Tamal K. Dey Jian Sun Yusu Wang Abstract Inference of topological and geometric attributes of a hidden manifold from
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 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 informationTOPOLOGY OF POINT CLOUD DATA
TOPOLOGY OF POINT CLOUD DATA CS 468 Lecture 8 11/12/2 Afra Zomorodian CS 468 Lecture 8 - Page 1 PROJECTS Writeups: Introduction to Knot Theory (Giovanni De Santi) Mesh Processing with Morse Theory (Niloy
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 informationCut Locus and Topology from Surface Point Data
Cut Locus and Topology from Surface Point Data Tamal K. Dey Kuiyu Li March 24, 2009 Abstract A cut locus of a point p in a compact Riemannian manifold M is defined as the set of points where minimizing
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 information1 Spaces and operations Continuity and metric spaces Topological spaces Compactness... 3
Compact course notes Topology I Fall 2011 Professor: A. Penskoi transcribed by: J. Lazovskis Independent University of Moscow December 23, 2011 Contents 1 Spaces and operations 2 1.1 Continuity and metric
More informationHomework in Topology, Spring 2009.
Homework in Topology, Spring 2009. Björn Gustafsson April 29, 2009 1 Generalities To pass the course you should hand in correct and well-written solutions of approximately 10-15 of the problems. For higher
More informationTOPOLOGY OF POINT CLOUD DATA
TOPOLOGY OF POINT CLOUD DATA CS 468 Lecture 8 3/3/4 Afra Zomorodian CS 468 Lecture 8 - Page 1 OVERVIEW Points Complexes Cěch Rips Alpha Filtrations Persistence Afra Zomorodian CS 468 Lecture 8 - Page 2
More informationarxiv: v1 [math.na] 10 Jun 2015
Continuation of Point Clouds via Persistence Diagrams Marcio Gameiro a, Yasuaki Hiraoka b, Ippei Obayashi b a Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, Caixa Postal
More information2.5 Excision implies Simplicial = Singular homology
2.5 Excision implies Simplicial = Singular homology 1300Y Geometry and Topology 2.5 Excision implies Simplicial = Singular homology Recall that simplicial homology was defined in terms of a -complex decomposition
More informationExercises for Algebraic Topology
Sheet 1, September 13, 2017 Definition. Let A be an abelian group and let M be a set. The A-linearization of M is the set A[M] = {f : M A f 1 (A \ {0}) is finite}. We view A[M] as an abelian group via
More information7. The Stability Theorem
Vin de Silva Pomona College CBMS Lecture Series, Macalester College, June 2017 The Persistence Stability Theorem Stability Theorem (Cohen-Steiner, Edelsbrunner, Harer 2007) The following transformation
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 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 informationCorrections to Introduction to Topological Manifolds (First edition) by John M. Lee December 7, 2015
Corrections to Introduction to Topological Manifolds (First edition) by John M. Lee December 7, 2015 Changes or additions made in the past twelve months are dated. Page 29, statement of Lemma 2.11: The
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 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 informationNeuronal structure detection using Persistent Homology
Neuronal structure detection using Persistent Homology J. Heras, G. Mata and J. Rubio Department of Mathematics and Computer Science, University of La Rioja Seminario de Informática Mirian Andrés March
More informationMath 147, Homework 5 Solutions Due: May 15, 2012
Math 147, Homework 5 Solutions Due: May 15, 2012 1 Let f : R 3 R 6 and φ : R 3 R 3 be the smooth maps defined by: f(x, y, z) = (x 2, y 2, z 2, xy, xz, yz) and φ(x, y, z) = ( x, y, z) (a) Show that f is
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 informationGromov s Proof of Mostow Rigidity
Gromov s Proof of Mostow Rigidity Mostow Rigidity is a theorem about invariants. An invariant assigns a label to whatever mathematical object I am studying that is well-defined up to the appropriate equivalence.
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 informationTree-adjoined spaces and the Hawaiian earring
Tree-adjoined spaces and the Hawaiian earring W. Hojka (TU Wien) Workshop on Fractals and Tilings 2009 July 6-10, 2009, Strobl (Austria) W. Hojka (TU Wien) () Tree-adjoined spaces and the Hawaiian earring
More informationDISCRETIZED CONFIGURATIONS AND PARTIAL PARTITIONS
DISCRETIZED CONFIGURATIONS AND PARTIAL PARTITIONS AARON ABRAMS, DAVID GAY, AND VALERIE HOWER Abstract. We show that the discretized configuration space of k points in the n-simplex is homotopy equivalent
More information5. 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 informationSolution: We can cut the 2-simplex in two, perform the identification and then stitch it back up. The best way to see this is with the picture:
Samuel Lee Algebraic Topology Homework #6 May 11, 2016 Problem 1: ( 2.1: #1). What familiar space is the quotient -complex of a 2-simplex [v 0, v 1, v 2 ] obtained by identifying the edges [v 0, v 1 ]
More informationFrom singular chains to Alexander Duality. Jesper M. Møller
From singular chains to Alexander Duality Jesper M. Møller Matematisk Institut, Universitetsparken 5, DK 21 København E-mail address: moller@math.ku.dk URL: http://www.math.ku.dk/~moller Contents Chapter
More informationALGEBRAIC PROPERTIES OF BIER SPHERES
LE MATEMATICHE Vol. LXVII (2012 Fasc. I, pp. 91 101 doi: 10.4418/2012.67.1.9 ALGEBRAIC PROPERTIES OF BIER SPHERES INGA HEUDTLASS - LUKAS KATTHÄN We give a classification of flag Bier spheres, as well as
More informationTowards Persistence-Based Reconstruction in Euclidean Spaces
INSTITUT NATIONAL DE RECHERCHE EN INFORMATIQUE ET EN AUTOMATIQUE arxiv:0712.2638v2 [cs.cg] 18 Dec 2007 Towards Persistence-Based Reconstruction in Euclidean Spaces Frédéric Chazal Steve Y. Oudot N 6391
More information6 Axiomatic Homology Theory
MATH41071/MATH61071 Algebraic topology 6 Axiomatic Homology Theory Autumn Semester 2016 2017 The basic ideas of homology go back to Poincaré in 1895 when he defined the Betti numbers and torsion numbers
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 informationChapter 3: Homology Groups Topics in Computational Topology: An Algorithmic View
Chapter 3: Homology Groups Topics in Computational Topology: An Algorithmic View As discussed in Chapter 2, we have complete topological information about 2-manifolds. How about higher dimensional manifolds?
More informationMATH 215B HOMEWORK 4 SOLUTIONS
MATH 215B HOMEWORK 4 SOLUTIONS 1. (8 marks) Compute the homology groups of the space X obtained from n by identifying all faces of the same dimension in the following way: [v 0,..., ˆv j,..., v n ] is
More informationCELLULAR HOMOLOGY AND THE CELLULAR BOUNDARY FORMULA. Contents 1. Introduction 1
CELLULAR HOMOLOGY AND THE CELLULAR BOUNDARY FORMULA PAOLO DEGIORGI Abstract. This paper will first go through some core concepts and results in homology, then introduce the concepts of CW complex, subcomplex
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 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 informationApplications of Sheaf Cohomology and Exact Sequences on Network Codings
Applications of Sheaf Cohomology and Exact Sequences on Network Codings Robert Ghrist Departments of Mathematics and Electrical & Systems Engineering University of Pennsylvania Philadelphia, PA, USA Email:
More informationConsistent Manifold Representation for Topological Data Analysis
Consistent Manifold Representation for Topological Data Analysis Tyrus Berry (tberry@gmu.edu) and Timothy Sauer Dept. of Mathematical Sciences, George Mason University, Fairfax, VA 3 Abstract For data
More informationarxiv: v3 [math.at] 27 Jan 2011
Optimal Homologous Cycles, Total Unimodularity, and Linear Programming Tamal K. Dey Anil N. Hirani Bala Krishnamoorthy arxiv:1001.0338v3 [math.at] 27 Jan 2011 Abstract Given a simplicial complex with weights
More informationMATH 215B HOMEWORK 5 SOLUTIONS
MATH 25B HOMEWORK 5 SOLUTIONS. ( marks) Show that the quotient map S S S 2 collapsing the subspace S S to a point is not nullhomotopic by showing that it induces an isomorphism on H 2. On the other hand,
More information3. Prove or disprove: If a space X is second countable, then every open covering of X contains a countable subcollection covering X.
Department of Mathematics and Statistics University of South Florida TOPOLOGY QUALIFYING EXAM January 24, 2015 Examiners: Dr. M. Elhamdadi, Dr. M. Saito Instructions: For Ph.D. level, complete at least
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 informationTopology Hmwk 6 All problems are from Allen Hatcher Algebraic Topology (online) ch 2
Topology Hmwk 6 All problems are from Allen Hatcher Algebraic Topology (online) ch 2 Andrew Ma August 25, 214 2.1.4 Proof. Please refer to the attached picture. We have the following chain complex δ 3
More informationGood Triangulations. Jean-Daniel Boissonnat DataShape, INRIA
Good Triangulations Jean-Daniel Boissonnat DataShape, INRIA http://www-sop.inria.fr/geometrica Algorithmic Geometry Good Triangulations J-D. Boissonnat 1 / 29 Definition and existence of nets Definition
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 informationHomology groups of neighborhood complexes of graphs (Introduction to topological combinatorics)
Homology groups of neighborhood complexes of graphs (Introduction to topological combinatorics) October 21, 2017 Y. Hara (Osaka Univ.) Homology groups of neighborhood complexes October 21, 2017 1 / 18
More informationDeclutter and Resample: Towards parameter free denoising
Declutter and Resample: Towards parameter free denoising Mickaël Buchet, Tamal K. Dey, Jiayuan Wang, Yusu Wang arxiv:1511.05479v2 [cs.cg] 27 Mar 2017 Abstract In many data analysis applications the following
More informationManifold Reconstruction in Arbitrary Dimensions using Witness Complexes
Manifold Reconstruction in Arbitrary Dimensions using Witness Complexes Jean-Daniel Boissonnat INRIA, Geometrica 2004, route des lucioles 06902 Sophia-Antipolis boissonn@sophia.inria.fr Leonidas J. Guibas
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 informationContents. 1. Introduction
DIASTOLIC INEQUALITIES AND ISOPERIMETRIC INEQUALITIES ON SURFACES FLORENT BALACHEFF AND STÉPHANE SABOURAU Abstract. We prove a new type of universal inequality between the diastole, defined using a minimax
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 informationMath 225B: Differential Geometry, Homework 8
Math 225B: Differential Geometry, Homewor 8 Ian Coley February 26, 204 Problem.. Find H (S S ) by induction on the number n of factors. We claim that H (T n ) ( n ). For the base case, we now that H 0
More informationDefinition 6.1. A metric space (X, d) is complete if every Cauchy sequence tends to a limit in X.
Chapter 6 Completeness Lecture 18 Recall from Definition 2.22 that a Cauchy sequence in (X, d) is a sequence whose terms get closer and closer together, without any limit being specified. In the Euclidean
More informationTHE PENNSYLVANIA STATE UNIVERSITY SCHREYER HONORS COLLEGE DEPARTMENT OF MATHEMATICS TOPOLOGY AND DATA ANALYSIS HANGYU ZHOU SPRING 2017
THE PENNSYLVANIA STATE UNIVERSITY SCHREYER HONORS COLLEGE DEPARTMENT OF MATHEMATICS TOPOLOGY AND DATA ANALYSIS HANGYU ZHOU SPRING 2017 A thesis submitted in partial fulfillment of the requirements for
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 informationReflections in Persistence and Quiver Theory
GT Combinatoire March 11, 2015 Reflections in Persistence and Quiver Theory Steve Oudot Geometrica group, INRIA Saclay Île-de-France 0-pers. Morse theory Clustering hierarchical mode analysis Spectral
More informationAlgebraic Topology exam
Instituto Superior Técnico Departamento de Matemática Algebraic Topology exam June 12th 2017 1. Let X be a square with the edges cyclically identified: X = [0, 1] 2 / with (a) Compute π 1 (X). (x, 0) (1,
More informationTOPOLOGY AND GROUPS. Contents
TOPOLOGY AND GROUPS Contents Introduction............................... 2 Convention and Notation......................... 3 I. Constructing spaces 1. Cayley graphs of groups..................... 4 2.
More informationNonabelian Poincare Duality (Lecture 8)
Nonabelian Poincare Duality (Lecture 8) February 19, 2014 Let M be a compact oriented manifold of dimension n. Then Poincare duality asserts the existence of an isomorphism H (M; A) H n (M; A) for any
More informationApproximating Riemannian Voronoi Diagrams and Delaunay triangulations.
Approximating Riemannian Voronoi Diagrams and Delaunay triangulations. Jean-Daniel Boissonnat, Mael Rouxel-Labbé and Mathijs Wintraecken Boissonnat, Rouxel-Labbé, Wintraecken Approximating Voronoi Diagrams
More informationMutiscale Mapper: A Framework for Topological Summarization of Data and Maps
Mutiscale Mapper: A Framework for Topological Summarization of Data and Maps Tamal K. Dey, Facundo Mémoli, Yusu Wang November 26, 2018 arxiv:1504.03763v1 [cs.cg] 15 Apr 2015 Abstract Summarizing topological
More informationSmith theory. Andrew Putman. Abstract
Smith theory Andrew Putman Abstract We discuss theorems of P. Smith and Floyd connecting the cohomology of a simplicial complex equipped with an action of a finite p-group to the cohomology of its fixed
More informationThe stability of Delaunay triangulations
The stability of Delaunay triangulations Jean-Daniel Boissonnat Ramsay Dyer Arijit Ghosh May 6, 2015 Abstract We introduce a parametrized notion of genericity for Delaunay triangulations which, in particular,
More informationG-invariant Persistent Homology
G-invariant Persistent Homology Patrizio Frosini, Dipartimento di Matematica, Università di Bologna Piazza di Porta San Donato 5, 40126, Bologna, Italy Abstract Classical persistent homology is not tailored
More information(iv) Whitney s condition B. Suppose S β S α. If two sequences (a k ) S α and (b k ) S β both converge to the same x S β then lim.
0.1. Stratified spaces. References are [7], [6], [3]. Singular spaces are naturally associated to many important mathematical objects (for example in representation theory). We are essentially interested
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 informationREFINMENTS OF HOMOLOGY. Homological spectral package.
REFINMENTS OF HOMOLOGY (homological spectral package). Dan Burghelea Department of Mathematics Ohio State University, Columbus, OH In analogy with linear algebra over C we propose REFINEMENTS of the simplest
More informationALGEBRAIC TOPOLOGY IV. Definition 1.1. Let A, B be abelian groups. The set of homomorphisms ϕ: A B is denoted by
ALGEBRAIC TOPOLOGY IV DIRK SCHÜTZ 1. Cochain complexes and singular cohomology Definition 1.1. Let A, B be abelian groups. The set of homomorphisms ϕ: A B is denoted by Hom(A, B) = {ϕ: A B ϕ homomorphism}
More informationTopological Analysis of Nerves, Reeb Spaces, Mappers, and Multiscale Mappers
Topological Analysis of Nerves, Reeb Spaces, Mappers, and Multiscale Mappers Tamal K. Dey, Facundo Mémoli, Yusu Wang arxiv:1703.07387v1 [cs.cg] 21 Mar 2017 Abstract Data analysis often concerns not only
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 informationCoxeter Groups and Artin Groups
Chapter 1 Coxeter Groups and Artin Groups 1.1 Artin Groups Let M be a Coxeter matrix with index set S. defined by M is given by the presentation: A M := s S sts }{{ } = tst }{{ } m s,t factors m s,t The
More informationTHE THEORY OF NORMAL SURFACES. based on Lecture Notes by Cameron McA. Gordon typeset by Richard P. Kent IV
THE THEORY OF NORMAL SURFACES based on Lecture Notes by Cameron McA. Gordon typeset by Richard P. Kent IV 1. Introduction When one considers questions like: When are two n-manifolds homeomorphic? When
More information1 Whitehead s theorem.
1 Whitehead s theorem. Statement: If f : X Y is a map of CW complexes inducing isomorphisms on all homotopy groups, then f is a homotopy equivalence. Moreover, if f is the inclusion of a subcomplex X in
More informationManifolds and Poincaré duality
226 CHAPTER 11 Manifolds and Poincaré duality 1. Manifolds The homology H (M) of a manifold M often exhibits an interesting symmetry. Here are some examples. M = S 1 S 1 S 1 : M = S 2 S 3 : H 0 = Z, H
More informationSimplicial Complexes in Finite Groups. (Cortona. July 2000)
Simplicial Complexes in Finite Groups (Cortona. July 2000) Firenze, September 2000. These notes contain the material presented at a Scuola Matematica Interuniversitaria summer course held in Cortona in
More informationTensor, Tor, UCF, and Kunneth
Tensor, Tor, UCF, and Kunneth Mark Blumstein 1 Introduction I d like to collect the basic definitions of tensor product of modules, the Tor functor, and present some examples from homological algebra 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 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 informationarxiv: v1 [math.co] 25 Jun 2014
THE NON-PURE VERSION OF THE SIMPLEX AND THE BOUNDARY OF THE SIMPLEX NICOLÁS A. CAPITELLI arxiv:1406.6434v1 [math.co] 25 Jun 2014 Abstract. We introduce the non-pure versions of simplicial balls and spheres
More informationMATH8808: ALGEBRAIC TOPOLOGY
MATH8808: ALGEBRAIC TOPOLOGY DAWEI CHEN Contents 1. Underlying Geometric Notions 2 1.1. Homotopy 2 1.2. Cell Complexes 3 1.3. Operations on Cell Complexes 3 1.4. Criteria for Homotopy Equivalence 4 1.5.
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 informationRELATIVE HYPERBOLICITY AND BOUNDED COHOMOLOGY
RELATIVE HYPERBOLICITY AND BOUNDED COHOMOLOGY IGOR MINEYEV AND ASLI YAMAN Abstract. Let Γ be a finitely generated group and Γ = {Γ i i I} be a family of its subgroups. We utilize the notion of tuple (Γ,
More informationEilenberg-Steenrod properties. (Hatcher, 2.1, 2.3, 3.1; Conlon, 2.6, 8.1, )
II.3 : Eilenberg-Steenrod properties (Hatcher, 2.1, 2.3, 3.1; Conlon, 2.6, 8.1, 8.3 8.5 Definition. Let U be an open subset of R n for some n. The de Rham cohomology groups (U are the cohomology groups
More informationAlgebraic Topology Lecture Notes. Jarah Evslin and Alexander Wijns
Algebraic Topology Lecture Notes Jarah Evslin and Alexander Wijns Abstract We classify finitely generated abelian groups and, using simplicial complex, describe various groups that can be associated to
More informationThe rational cohomology of real quasi-toric manifolds
The rational cohomology of real quasi-toric manifolds Alex Suciu Northeastern University Joint work with Alvise Trevisan (VU Amsterdam) Toric Methods in Homotopy Theory Queen s University Belfast July
More informationOn Inverse Problems in TDA
Abel Symposium Geiranger, June 2018 On Inverse Problems in TDA Steve Oudot joint work with E. Solomon (Brown Math.) arxiv: 1712.03630 Features Data Rn The preimage problem in the data Sciences feature
More information