Some Topics in Computational Topology
|
|
- Winifred Davidson
- 5 years ago
- Views:
Transcription
1 Some Topics in Computational Topology Yusu Wang Ohio State University AMS Short Course 2014
2 Introduction Much recent developments in computational topology Both in theory and in their applications E.g, the theory of persistence homology [Edelsbrunner, Letscher, Zomorodian, DCG 2002], [Zomorodian and Carlsson, DCG 2005], [Carlsson and de Silva, FoCM 2010], This short course: A computational perspective: Estimation and inference of topological information / structure from point clouds data
3 ? Topological summary of hidden space Unorganized PCD Develop discrete analog for the continuous case Approximation from discrete samples with theoretical guarantees Algorithmic issues
4 Main Topics From PCDs to simplicial complexes Sampling conditions Topological inferences
5 Outline From PCDs to simplicial complexes Delaunay, Cech, Vietoris-Rips, witness complexes Graph induced complex Sampling conditions Local feature size, and homological feature size Topology inference Homology inference Handling noise Approximating cycles of shortest basis of the first homology group Approximating Reeb graph
6 From PCD to Simplicial Complexes Choice of Simplicial Complexes to build on top of point cloud data
7 Delaunay Complex Given a set of points P = p 1, p 2,, p n R d Delaunay complex Del P A simplex σ = {p i0, p i1,, p i k } is in Del P if and only if There exists a ball B whose boundary contains vertices of σ, and that the interior of B contains no other point from P.
8 Delaunay Complex Many beautiful properties Connection to Voronoi diagram Foundation for surface reconstruction and meshing in 3D [Dey, Curve and Surface Reconstruction, 2006], [Cheng, Dey and Shewchuk, Delaunay Mesh Generation, 2012] However, Computationally very expensive in high dimensions
9 Čech Complex Given a set of points P = p 1, p 2,, p n R d Given a real value r > 0, the Čech complex C r P is the nerve of the set B p i, r i 1,n where B r p i = B p i, r = { x R d d p i, x < r } I.e, a simplex σ = {p i0, p i1,, p i k } is in Cr P if 0 j k B r p ij Φ. The definition can be extended to a finite sample P of a metric space.
10 Rips Complex Given a set of points P = p 1, p 2,, p n R d Given a real value r > 0, the Vietoris-Rips (Rips) complex R r P is: { p i0, p i1,, p ik B r p il B r p ij, l, j [0, k ]}.
11 Rips and Čech Complexes Relation in general metric spaces C r P R r P C 2r P Bounds better in Euclidean space Simple to compute Able to capture geometry and topology We will make it precise shortly One of the most popular choices for topology inference in recent years However: Huge sizes Computation also costly
12 Witness Complexes A simplex σ = q 0,, q k is weakly witnessed by a point x if d q i, x d q, x for any i 0, k and q Q {q 0,, q k }. is strongly witnessed if in addition d q i, x = d q j, x, i, j 0, k Given a set of points P = p 1, p 2,, p n R d and a subset Q P, The witness complex W Q, P is the collection of simplices with vertices from Q whose all subsimplices are weakly witnessed by a point in P. [de Silva and Carlsson, 2004] [de Silva 2003] Can be defined for a general metric space P does not have to be a finite subset of points
13 Intuition L: landmarks from P, a way to subsample. P L P W(L, P)
14 Witness Complexes Greatly reduce size of complex Similar to Delaunay triangulation, remove redundancy Relation to Delaunay complex W Q, P Del Q if Q P R d W Q, R d = Del Q W Q, M [Attali et al, 2007] = Del M Q if M R d is a smooth 1- or 2-manifold However, Does not capture full topology easily for high-dimensional manifolds
15 Remark Rips complex Capture homology when input points are sampled dense enough But too large in size Witness complex Use a subsampling idea Reduce size tremendously May not be easy to capture topology in high-dimensions Combine the two? Graph induced complex [Dey, Fan, Wang, SoCG 2013]
16 Subsampling P R ε (P)
17 Subsampling -cont P Q P R r Q W(Q, P)
18 Subsampling -cont Q P R r Q W(Q, P)
19 Subsampling - cont R ε (P) W(Q, P) G r (Q, P)
20 Graph Induced Complex [Dey, Fan, Wang, SoCG 2013] P: finite set of points (P, d): metric space G(P): a graph Q P: a subset π p : the closest point of p P in Q
21 Graph Induced Complex Graph induced complex G P, Q, d : q 0,, q k Q if and only if there is a (k+1)-clique in G P with vertices p 0,, p k such that π p i = q i, for any i [0, k]. Similar to geodesic Delaunay [Oudot, Guibas, Gao, Wang, 2010] Graph induced complex depends on the metric d: Euclidean metric Graph based distance d G
22 Graph Induced Complex Small size, but with homology inference guarantees In particular: H 1 inference from a lean sample
23 Graph Induced Complex Small size, but with homology inference guarantees In particular: H 1 inference from a lean sample Surface reconstruction in R 3 Topological inference for compact sets in R d using persistence
24 Outline From PCDs to simplicial complexes Delaunay, Cech, Vietoris-Rips, witness complexes Graph induced complex Sampling conditions Local feature size, and homological feature size Topology inference Homology inference Handling noise Approximating cycles of shortest basis of the first homology group Approximating Reeb graph
25 Sampling Conditions
26 Motivation Theoretical guarantees are usually obtained when input points P sampling the hidden domain well enough. Need to quantify the wellness. Two common ones based on: Local feature size Weak feature size
27 Distance Function X R d : a compact subset of R d Distance function d X : R d R + 0 d X x = min y X d x, y d X is a 1-Lipschitz function X α : α-offset of X X α = {y R d d y, X α} Given any point x R d Γ x { y X d x, y = d X x }
28 Medial Axis The medial axis Σ of X is the closure of the set of points x R d such that Γ x 2 Γ x 2 means that there is a medial ball B r x touching X at more than 1 point and whose interior is empty of points from X. Courtesy of [Dey, 2006]
29 Local Feature Size The local feature size lfs x at a point x X is the distance of x to the medial axis Σ of X That is, lfs x = d x, Σ This concept is adaptive Large in a place without features Intuitively: We should sample more densely if local feature size is small. The reach ρ X = inf x X lfs x Courtesy of [Dey, 2006]
30 Gradient of Distance Function Distance function not differentiable on the medial axis Still can define a generalized concept of gradient [Lieutier, 2004] For x R d X, Let c X x and r X x be the center and radius of the smallest enclosing ball of point(s) in Γ x The generalized gradient of distance function X x = x c X x r X x Flow lines induced by the generalized gradient Examples:
31 Critical Points A critical point of the distance function is a point whose generalized gradient X x vanishes A critical point is either in X or in its medial axis Σ
32 Weak Feature Size Given a compact X R d, let C R d denote the set of critical points of the distance function d X that are not in X Given a compact X R d, the weak feature size is wfs X Equivalently, = inf d x, C x X wfs X is the infimum of the positive critical value of d X ρ X wfs X
33 Why Distance Field? Theorem [Offset Homotopy] [Grove 93] If 0 < α < α are such that there is no critical value of d X in the closed interval α, α, then X α deformation retracts onto X α. In particular, H X α H X α. Remarks: For the case of compact set X, note that it is possible that X α, for sufficiently small α > 0, may not be homotopy equivalent to X 0 = X. Intuitively, by above theorem, we can approximate H X α for any small positive α from a thickened version (offset) of X α. The sampling condition makes sure that the discrete sample is sufficient to recover the offset homology.
34 Typical Sampling Conditions Hausdorff distance d H A, B between two sets A and B infinumum value α such that A B α and B A α No noise version: A set of points P is an ε-sample of X if P X and d H P, X ε With noise version: A set of points P is an ε-sample of X if d H P, X ε Theoretical guarantees will be achieved when ε is small with respect to local feature size or weak feature size
35 Outline From PCDs to simplicial complexes Delaunay, Cech, Vietoris-Rips, witness complexes Graph induced complex Sampling conditions Local feature size, and homological feature size Topology inference Homology inference Handling noise Approximating cycles of shortest basis of the first homology group Approximating Reeb graph
36 Homology Inference from PCD
37 Problem Setup A hidden compact X (or a manifold M) An ε-sample P of X Recover homology of X from some complex built on P will focus on Čech complex and Rips complex
38 Union of Balls X α = P α = x X p P B x, α B p, α Intuitively, P α approximates offset X α The Čech complex C α P is the Nerve of P α By Nerve Lemma, C α P is homotopy equivalent to P α
39 Smooth Manifold Case Let X be a smooth manifold embedded in R d Theorem [Niyogi, Smale, Weinberger] Let P X be such that d H X, P ε. If 2ε α 3 5 there is a deformation retraction from P α to X. ρ X, Corollary A Under the conditions above, we have H X H P α H C α P.
40 How about using Rips complex instead of Čech complex? Recall that inducing C r P R r P C 2r P H(C r ) H(R r ) H(C 2r ) Idea [Chazal and Oudot 2008]: Forming interleaving sequence of homomorphism to connect them with the homology of the input manifold X and its offsets X α
41 Convert to Cech Complexes Lemma A [Chazal and Oudot, 2008]: The following diagram commutes: i H P α H P β h h i H C α H C β Corollary B Let P X be s.t. d H X, P ε. If 2ε α α 3 5 ρ X, H X H C α H C β where the second isomorphism is induced by inclusion.
42 From Rips Complex Lemma B: Given a sequence A B C D E of homomorphisms between finite dimensional vector spaces, if rank A E = rank C, then rank B D = dim C. Rips and Čech complexes: C α P R α P C 2α P R 2α P C 4α P H(C α ) H(R α ) H C 2α H R 2α H C 4α Applying Lemma B rank H R α H R 2α = rank H C 2α = rank H X
43 The Case of Compact In contrast to Corollary A, now we have the following (using Lemma B). Lemma C [Chazal and Oudot 2008]: Let P R d be a finite set such that d H X, P < ε for some ε < 1 wfs X. Then for all α, β ε, wfs X ε such 4 that β α 2ε, and for all λ 0, wfs X, we have H X λ image i, where i : H P α H P β is the homomorphism between homology groups induced by the canonical inclusion i: P α P β.
44 The Case of Compacts One more level of interleaving. Use the following extension of Lemma B: Given a sequence A B C D E F of homomorphisms between finite dimensional vector spaces, if rank A F = rank C D, then rank B E = rank C D. Theorem [Homology Inference] [Chazal and Oudot 2008]: Let P R d be a finite set such that d H X, P < ε for some ε < 1 wfs X. Then for all α 2ε, 1 wfs X ε all 9 4 λ 0, wfs X, we have H X λ image j, where j is the homomorphism between homology groups induced by canonical inclusion j: R α R 4α.
45 Theorem [Homology Inference] [Chazal and Oudot 2008]: Let P R d be a finite set such that d H X, P < ε for some ε < 1 wfs X. Then for all α 2ε, 1 wfs X ε all λ 9 4 0, wfs X, we have H X λ image j, where j is the homomorphism between homology groups induced by canonical inclusion j: R α R 4α.
46 Summary of Homology Inference X α homotopy equivalent to X β Critical points of distance field P α approximates X α (may be interleaving) E.g, [Niyogi, Smale, Weinberger, 2006], [Chazal and Oudot 2008] C α homotopy equivalent to P α Nerve Lemma H(C α ) interleaves H(X α ) at homology level [Chazal and Oudot 2008] R α and C α interleave Derive homology inference from the interleaving sequence of homomorphisms
47 Outline From PCDs to simplicial complexes Delaunay, Cech, Vietoris-Rips, witness complexes Graph induced complex Sampling conditions Local feature size, and homological feature size Topology inference Homology inference Handling noise Approximating cycles of shortest basis of the first homology group Approximating Reeb graph
48 Handling of Noise
49 Noise Previous approaches can handle Hausdorff type noise Where noise is within a tubular neighborhood of X How about more general noise? E.g, Gaussian noise, background noise Some approaches Bound the probability that input samples fall in Hausdorff model Denoise so that resulting points fall in Hausdorff model E.g, [Niyogi, Smale, Weinberger 2006, 2008] Distance to measure framework [Chazal, Cohen-Steiner, Mérigot, 2011]
50 Overview Input: A set of points P sampled from a probabilistic measure μ on R d potentially concentrated on a hidden compact (e.g, manifold) X. Goal: Approximate topological features of X Courtesy of Chazal et al 2011
51 Main Idea The work of [Chazal, Cohen-Steiner, Mérigot, 2011] A new distance function d μ,m to a probability μ measure (ie., distance to measures) to replace the role of distance function d X. Show that the two distance fields are close (in L norm) Topological inference follows from some stability results or the interleaving sequences
52 Definitions μ: a probability measure on R d ; μ R d = 1 0 < m, m 0 < 1: mass parameters δ μ,m : a pseudo-distance function such that δ μ,m x inf{r > 0; μ B x, r > m } where B x, r is the closed Euclidean ball at x That is, δ μ,m x is the radius of the ball necessary in order to enclose mass m Distance to measure d μ,m0 : m 0 d 2 μ,m 0 x = 1 m 0 δ μ,m x 2 dm 0
53 Distance to Measures Distance to measure d μ,m0 : 2 d μ,m0 x = 1 δ m μ,m x 2 dm 0 0 Intuitive, d μ,m x averages distance within a range and is more robust to noise. Note this distance depends on a mass parameter m 0 m 0
54 Wasserstein Distance A transport plan between two probability measures μ, ν on R d is a probability measure π on R d R d such that for every A, B R d, π A R d = μ A and π R d B = μ B. The p-cost of a transport plan π is: C p π = x y p dπ x, y R d R d The Wasserstein distance of order p between μ, ν on R d with finite p-moment W p μ, ν = the minimum p-cost C p π of any transport plan π between μ and ν. 1 p
55 Properties Theorem [Distance-Likeness] [Chazal et al 2011] 2 d μ,m0 is distance like. That is: The function d μ,m0 is 1-Lipschitz. 2 The function d μ,m0 is 1-semiconcave, meaning that the map 2 x d μ,m0 x x 2 is concave. Theorem [Stability] [Chazal et al 2011] Let μ, μ be two probability measures on R d and m 0 > 0. Then d μ,m0 d μ,m 1 0 W m 2 (μ, μ ) 0
56 Distance-Like Function There are many interesting consequences of [Theorem Distance-Likeness Theorem]. E.g, one can define critical points, weak feature size, etc. Theorem [Isotopy Lemma]: Let φ be a distance-like function and r 1 < r 2 be two positive numbers such that φ has no critical points in the subset φ 1 r 1, r 2. Then all the sublevel sets φ 1 0, r are isotopic for r [r 1, r 2 ].
57 Relation to Distance Function Now suppose P is sampled from, not compact X, but a probabilistic measure μ on R d concentrated X. Consider P as a noisy sample of X Let ν X denote the uniform measure on X Theorem [Approximation Distance]: d X d μ,m0 C X 1 km 0 1 k + 1 m 0 W 2 μ, ν X where X is a k-dimensional smooth manifold and C(X) is a quantity depending on X and k.
58 Computational Aspect Distance to measures can be approximated efficiently for a set of points P [Guibas, Mérigot and Morozov, DCG 2013] Can be extended to metric spaces, and build weighted Cech / Rips complexes for reconstruction and homology inference [Buchet, Chazal, Oudot, Sheehy, 2013]
59 Outline From PCDs to simplicial complexes Delaunay, Cech, Vietoris-Rips, witness complexes Graph induced complex Sampling conditions Local feature size, and homological feature size Topology inference Homology inference Handling noise Approximating cycles of shortest basis of the first homology group Approximating Reeb graph
60 Two Additional Examples Previously: Homology inference: topological information of a space Approximating cycles of shortest basis of the first homology group As an example of combining geometry and topology Approximating Reeb graph As an example of approximating the topology of a scalar field See separate slides for these two additional topics.
61 Summary One example of the pipeline of approximating certain topological structure from discrete samples The components in the pipeline are quite generic Many other issues too: Stability Efficiency Sparsification etc
62 Summary cont. Starting to have more interaction with statistics and probability theory E.g, [Balakrishnan et al., AISTATS 2012], [Bendich et al., SoDA 2012] How to develop algorithms that integrating computational geometry / topology ideas with statistics, especially in data analysis?
63 References Weak witnesses for Delaunay triangulations of submanifolds. D. Attali, H. Edelsbrunner, and Y. Mileyko. Proc. ACM Sympos. on Solid and Physical Modeling, (2007) Manifold reconstruction in arbitrary dimensions using witness complexes. D. Boissonnat and L. J. Guibas and S. Y. Oudot. SoCG , (2007) Efficient and Robust Topological Data Analysis on Metric Spaces. M. Buchet, F. Chazal, S. Y. Oudot, D. Sheehy:. CoRR abs/ (2013) Geometric Inference. F. Chazal and D. Cohen-Steiner. Tesselations in the Sciences, Springer-Verlag, (2013, to appear). A sampling theory for compacts in Euclidean space. F. Chazal, D. Cohen-Steiner, and A. Lieutier. Discrete Comput. Geom. 41, (2009). Geometric Inference for Probability Measures. F. Chazal, D. Cohen-Steiner, Q. Mérigot. Foundations of Computational Mathematics 11(6): (2011) Stability and Computation of Topological Invariants of Solids in R n. F. Chazal, A. Lieutier. Discrete & Computational Geometry 37(4): (2007)
64 References cont. Towards persistence-based reconstruction in Euclidean spaces. F. Chazal and S. Oudot. Proc. Annu. Sympos. Comput. Geom. (2008), Proximity of persistence modules and their diagrams. F. Chazal, D. Cohen- Steiner, M. Glisse, L. Guibas, and S. Oudot. Proc. 25th Annu. Sympos. Comput. Geom. (2009), A weak definition of Delaunay triangulation. Vin de Silva. CoRR cs.cg/ (2003) Topological estimation using witness complexes. V. de Silva, G. Carlsson, Symposium on Point-Based Graphics,, 2004 Topological estimation using witness complexes. Vin de Silva and Gunnar Carlsson. Eurographics Sympos. Point-Based Graphics (2004). Curve and Surface Reconstruction : Algorithms with Mathematical Analysis. Tamal. K. Dey. Cambridge University Press, Graph induced complex on point data. T. K. Dey, F. Fan, and Y. Wang. to appear in 29th Annual Sympos. Comput. Geom. (SoCG) 2013.
65 References cont. Witnessed k-distance. L. J. Guibas, D. Morozov, Q. Mérigot. Discrete & Computational Geometry 49(1): (2013) Reconstructing using witness complexes, L. J. Guibas and S. Y. Oudot. Discrete Comput. Geom. 30, (2008). Any open bounded subset of R n has the same homotopy type as its medial axis. A. Lieutier. Computer-Aided Design 36(11): (2004) Finding the homology of submanifolds with high confidence from random samples. P. Niyogi, S. Smale, and S. Weinberger. Discrete Comput. Geom., 39, (2008). Geodesic delaunay triangulations in bounded planar domains. S. Oudot, L. J. Guibas, J. Gao and Y. Wang. ACM Trans. Alg. 6 (4): 2010.
66 References cont. Minimax rates for homology inference. S. Balakrishnan, A. Rinaldo, D. Sheehy, A. Singh, L. A. Wasserman. AISTATS 2012: Local Homology Transfer and Stratification Learning. P. Bendich, B. Wang and S. Mukherjee. ACM-SIAM Symposium on Discrete Algorithms, (2012). Greedy optimal homotopy and homology generators. J. Erickson and K. Whittlesey. Proc. Sixteenth ACM-SIAM Sympos. Discrete Algorithms (2005), Approximating cycles in a shortest basis of the first homology group from point data. T. K. Dey, J. Sun, and Y. Wang. Inverse Problems, 27 (2011). Reeb graphs: Approximation and Persistence. T. K. Dey and Y. Wang., Discrete Comput. Geom (DCG) 2013.
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 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 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 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 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 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 June 1 Motivation What is the (relevant)
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 informationA sampling theory for compact sets
ENS Lyon January 2010 A sampling theory for compact sets F. Chazal Geometrica Group INRIA Saclay To download these slides: http://geometrica.saclay.inria.fr/team/fred.chazal/teaching/distancefunctions.pdf
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 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 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 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 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 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 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 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 informationWITNESSED K-DISTANCE
WITNESSED K-DISTANCE LEONIDAS GUIBAS, DMITRIY MOROZOV, AND QUENTIN MÉRIGOT Abstract. Distance function to a compact set plays a central role in several areas of computational geometry. Methods that rely
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. 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 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 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 informationStability and computation of topological invariants of solids in R n
Stability and computation of topological invariants of solids in R n Frédéric Chazal André Lieutier Abstract In this work, one proves that under quite general assumptions one can deduce the topology of
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 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 informationGeometric Inference for Probability Measures
Geometric Inference for Probability Measures Frédéric CHAZAL David COHEN-STEINER Quentin MÉRIGOT June 21, 2010 Abstract Data often comes in the form of a point cloud sampled from an unknown compact subset
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 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 informationWeak Witnesses for Delaunay Triangulations of Submanifolds
Weak Witnesses for Delaunay Triangulations of Submanifolds Dominique Attali LIS, INPG-CNRS Herbert Edelsbrunner Duke University and Geomagic Yuriy Mileyko Duke University Abstract The main result of this
More informationTopology Guaranteeing Manifold Reconstruction using Distance Function to Noisy Data
Topology Guaranteeing Manifold Reconstruction using Distance Function to Noisy Data ABSTRACT Frédéric Chazal Université de Bourgogne Institute of Mathematics of Burgundy France fchazal@u-bourgogne.fr Given
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 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 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 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 Local Homology: Theory, Applications, Algorithms
Persistent Local Homology: Theory, Applications, Algorithms Paul Bendich Duke University Feb. 4, 2014 Basic Goal: stratification learning Q: Given a point cloud sampled from a stratified space, how do
More informationStatistics and Topological Data Analysis
Geometry Understanding in Higher Dimensions CHAIRE D INFORMATIQUE ET SCIENCES NUMÉRIQUES Collège de France - June 2017 Statistics and Topological Data Analysis Bertrand MICHEL Ecole Centrale de Nantes
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 informationMinimax Rates. Homology Inference
Minimax Rates for Homology Inference Don Sheehy Joint work with Sivaraman Balakrishan, Alessandro Rinaldo, Aarti Singh, and Larry Wasserman Something like a joke. Something like a joke. What is topological
More informationGeometric inference for measures based on distance functions The DTM-signature for a geometric comparison of metric-measure spaces from samples
Distance to Geometric inference for measures based on distance functions The - for a geometric comparison of metric-measure spaces from samples the Ohio State University wan.252@osu.edu 1/31 Geometric
More informationGromov-Hausdorff stable signatures for shapes using persistence
Gromov-Hausdorff stable signatures for shapes using persistence joint with F. Chazal, D. Cohen-Steiner, L. Guibas and S. Oudot Facundo Mémoli memoli@math.stanford.edu l 1 Goal Shape discrimination is a
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 informationWitnessed k-distance
Witnessed k-distance Leonidas J. Guibas, Quentin Mérigot, Dmitriy Morozov To cite this version: Leonidas J. Guibas, Quentin Mérigot, Dmitriy Morozov. Witnessed k-distance. Discrete and Computational Geometry,
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 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 informationStability of boundary measures
Stability of boundary measures F. Chazal D. Cohen-Steiner Q. Mérigot INRIA Saclay - Ile de France LIX, January 2008 Point cloud geometry Given a set of points sampled near an unknown shape, can we infer
More informationWe have the following immediate corollary. 1
1. Thom Spaces and Transversality Definition 1.1. Let π : E B be a real k vector bundle with a Euclidean metric and let E 1 be the set of elements of norm 1. The Thom space T (E) of E is the quotient E/E
More informationSmooth manifold reconstruction from noisy and non-uniform approximation with guarantees
Computational Geometry 40 2008) 156 170 www.elsevier.com/locate/comgeo Smooth manifold reconstruction from noisy and non-uniform approximation with guarantees Frédéric Chazal a,, André Lieutier b a INRIA
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 informationLearning from persistence diagrams
Learning from persistence diagrams Ulrich Bauer TUM July 7, 2015 ACAT final meeting IST Austria Joint work with: Jan Reininghaus, Stefan Huber, Roland Kwitt 1 / 28 ? 2 / 28 ? 2 / 28 3 / 28 3 / 28 3 / 28
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 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 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 informationAlpha-Shapes and Flow Shapes are Homotopy Equivalent
Alpha-Shapes and Flow Shapes are Homotopy Equivalent Tamal K. Dey The Ohio State University, USA tamaldey@cis.ohio-state.edu Joachim Giesen ETH Zürich, Switzerland giesen@inf.ethz.ch Matthias John ETH
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 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 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 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 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 informationPersistence Theory: From Quiver Representations to Data Analysis
Persistence Theory: From Quiver Representations to Data Analysis Comments and Corrections Steve Oudot June 29, 2017 Comments page viii, bottom of page: the following names should be added to the acknowledgements:
More informationApproximating the Pathway Axis and the Persistence Diagram of a Collection of Balls in 3-Space
Approximating the Pathway Axis and the Persistence Diagram of a Collection of Balls in 3-Space Eitan Yaffe Dan Halperin School of Computer Science Tel-Aviv University, Israel {eitanyaf,danha}@post.tau.ac.il
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 information32 Proof of the orientation theorem
88 CHAPTER 3. COHOMOLOGY AND DUALITY 32 Proof of the orientation theorem We are studying the way in which local homological information gives rise to global information, especially on an n-manifold M.
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 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 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 informationComputing the Robustness of Roots
Computing the Robustness of Roots Frédéric Chazal frederic.chazal@inria.fr Amit Patel amit.patel@inria.fr Primoz Skraba primoz.skraba@inria.fr Abstract Often relationships in nature are modelled as mappings
More informationNeural Codes and Neural Rings: Topology and Algebraic Geometry
Neural Codes and Neural Rings: Topology and Algebraic Geometry Ma191b Winter 2017 Geometry of Neuroscience References for this lecture: Curto, Carina; Itskov, Vladimir; Veliz-Cuba, Alan; Youngs, Nora,
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 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 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 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 informationGromov-Hausdorff Approximation of Metric Spaces with Linear Structure
Gromov-Hausdorff Approximation of Metric Spaces with Linear Structure Frédéric Chazal, Jian Sun To cite this version: Frédéric Chazal, Jian Sun. Gromov-Hausdorff Approximation of Metric Spaces with Linear
More informationTopological Data Analysis with Bregman Divergences
Topological Data Analysis with Bregman Divergences Herbert Edelsbrunner 1 and Hubert Wagner 2 1 IST Austria, Klosterneuburg, Austria edels@ist.ac.at 2 IST Austria, Klosterneuburg, Austria hub.wag@gmail.com
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 informationTransversality. Abhishek Khetan. December 13, Basics 1. 2 The Transversality Theorem 1. 3 Transversality and Homotopy 2
Transversality Abhishek Khetan December 13, 2017 Contents 1 Basics 1 2 The Transversality Theorem 1 3 Transversality and Homotopy 2 4 Intersection Number Mod 2 4 5 Degree Mod 2 4 1 Basics Definition. Let
More informationBeyond random graphs : random simplicial complexes. Applications to sensor networks
Beyond random graphs : random simplicial complexes. Applications to sensor networks L. Decreusefond E. Ferraz P. Martins H. Randriambololona A. Vergne Telecom ParisTech Workshop on random graphs, April
More informationBredon, Introduction to compact transformation groups, Academic Press
1 Introduction Outline Section 3: Topology of 2-orbifolds: Compact group actions Compact group actions Orbit spaces. Tubes and slices. Path-lifting, covering homotopy Locally smooth actions Smooth actions
More informationHeat Kernel Signature: A Concise Signature Based on Heat Diffusion. Leo Guibas, Jian Sun, Maks Ovsjanikov
Heat Kernel Signature: A Concise Signature Based on Heat Diffusion i Leo Guibas, Jian Sun, Maks Ovsjanikov This talk is based on: Jian Sun, Maks Ovsjanikov, Leonidas Guibas 1 A Concise and Provably Informative
More information2 Review: Modeling Molecules with Spherical Balls
CS273: Algorithms for Structure Handout # 12 and Motion in Biology Stanford University Thursday, 6 May 2003 Lecture #12: 6 May 2004 Topics: Geometric Models of Molecules II Scribe: Itamar Rosenn 1 Introduction
More informationGlobal (ISOMAP) versus Local (LLE) Methods in Nonlinear Dimensionality Reduction
Global (ISOMAP) versus Local (LLE) Methods in Nonlinear Dimensionality Reduction A presentation by Evan Ettinger on a Paper by Vin de Silva and Joshua B. Tenenbaum May 12, 2005 Outline Introduction The
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 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 informationCutting and pasting. 2 in R. 3 which are not even topologically
Cutting and pasting We begin by quoting the following description appearing on page 55 of C. T. C. Wall s 1960 1961 Differential Topology notes, which available are online at http://www.maths.ed.ac.uk/~aar/surgery/wall.pdf.
More informationOn the Diffeomorphism Group of S 1 S 2. Allen Hatcher
On the Diffeomorphism Group of S 1 S 2 Allen Hatcher This is a revision, written in December 2003, of a paper of the same title that appeared in the Proceedings of the AMS 83 (1981), 427-430. The main
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 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 informationThe Intersection of Statistics and Topology:
The Intersection of Statistics and Topology: Confidence Sets for Persistence Diagrams Brittany Terese Fasy joint work with S. Balakrishnan, F. Lecci, A. Rinaldo, A. Singh, L. Wasserman 3 June 2014 [FLRWBS]
More informationA NOTE ON SPACES OF ASYMPTOTIC DIMENSION ONE
A NOTE ON SPACES OF ASYMPTOTIC DIMENSION ONE KOJI FUJIWARA AND KEVIN WHYTE Abstract. Let X be a geodesic metric space with H 1(X) uniformly generated. If X has asymptotic dimension one then X is quasi-isometric
More informationGromov-Hausdorff Approximation of Filament Structure Using Reeb-type Graph
Gromov-Hausdorff Approximation of Filament Structure Using Reeb-type Graph Frédéric Chazal and Ruqi Huang and Jian Sun July 26, 2014 Abstract In many real-world applications data appear to be sampled around
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 informationGromov-Hausdorff Approximation of Filament Structure Using Reeb-type Graph
Noname manuscript No. (will be inserted by the editor) Gromov-Hausdorff Approximation of Filament Structure Using Reeb-type Graph Frédéric Chazal Ruqi Huang Jian Sun Received: date / Accepted: date Abstract
More informationFinding the Homology of Submanifolds with High Confidence from Random Samples
Finding the Homology of Submanifolds with High Confidence from Random Samples P. Niyogi, S. Smale, S. Weinberger March 15, 2006 Abstract Recently there has been a lot of interest in geometrically motivated
More information121B: ALGEBRAIC TOPOLOGY. Contents. 6. Poincaré Duality
121B: ALGEBRAIC TOPOLOGY Contents 6. Poincaré Duality 1 6.1. Manifolds 2 6.2. Orientation 3 6.3. Orientation sheaf 9 6.4. Cap product 11 6.5. Proof for good coverings 15 6.6. Direct limit 18 6.7. Proof
More informationGraphs Representations and the Topology of Real and Angle valued maps (an alternative approach to Morse-Novikiov theory)
Graphs Representations and the Topology of Real and Angle valued maps (an alternative approach to Morse-Novikiov theory) Dan Burghelea Department of mathematics Ohio State University, Columbus, OH Tianjin,
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 informationMath 6510 Homework 10
2.2 Problems 9 Problem. Compute the homology group of the following 2-complexes X: a) The quotient of S 2 obtained by identifying north and south poles to a point b) S 1 (S 1 S 1 ) c) The space obtained
More informationarxiv: v2 [math.pr] 15 May 2016
MAXIMALLY PERSISTENT CYCLES IN RANDOM GEOMETRIC COMPLEXES OMER BOBROWSKI, MATTHEW KAHLE, AND PRIMOZ SKRABA arxiv:1509.04347v2 [math.pr] 15 May 2016 Abstract. We initiate the study of persistent homology
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 informationComputational Homology in Topological Dynamics
1 Computational Homology in Topological Dynamics ACAT School, Bologna, Italy MAY 26, 2012 Marian Mrozek Jagiellonian University, Kraków Outline 2 Dynamical systems Rigorous numerics of dynamical systems
More information