1 By: MOSES CHARIKAR, CHANDRA CHEKURI, TOMAS FEDER, AND RAJEEV MOTWANI Presented By: Sarah Hegab.

Slides:



Advertisements
Similar presentations
Routing Complexity of Faulty Networks Omer Angel Itai Benjamini Eran Ofek Udi Wieder The Weizmann Institute of Science.
Advertisements

Chapter 5: Tree Constructions
WSPD Applications.
Clustering.
Hierarchical Clustering. Produces a set of nested clusters organized as a hierarchical tree Can be visualized as a dendrogram – A tree-like diagram that.
1 SOFSEM 2007 Weighted Nearest Neighbor Algorithms for the Graph Exploration Problem on Cycles Eiji Miyano Kyushu Institute of Technology, Japan Joint.
Approximation Algorithms Chapter 14: Rounding Applied to Set Cover.
Incremental Clustering Previous clustering algorithms worked in “batch” mode: processed all points at essentially the same time. Some IR applications cluster.
Heaps1 Part-D2 Heaps Heaps2 Recall Priority Queue ADT (§ 7.1.3) A priority queue stores a collection of entries Each entry is a pair (key, value)
Greedy Algorithms Greed is good. (Some of the time)
Approximation, Chance and Networks Lecture Notes BISS 2005, Bertinoro March Alessandro Panconesi University La Sapienza of Rome.
Study Group Randomized Algorithms 21 st June 03. Topics Covered Game Tree Evaluation –its expected run time is better than the worst- case complexity.
Bounds on Code Length Theorem: Let l ∗ 1, l ∗ 2,..., l ∗ m be optimal codeword lengths for a source distribution p and a D-ary alphabet, and let L ∗ be.
Dynamic Wavelength Allocation in All-optical Ring Networks Ori Gerstel and Shay Kutten Proceedings of ICC'97.
Robust hierarchical k- center clustering Ilya Razenshteyn (MIT) Silvio Lattanzi (Google), Stefano Leonardi (Sapienza University of Rome) and Vahab Mirrokni.
Approximation Algorithms Chapter 5: k-center. Overview n Main issue: Parametric pruning –Technique for approximation algorithms n 2-approx. algorithm.
Combinatorial Algorithms
Optimization of Pearl’s Method of Conditioning and Greedy-Like Approximation Algorithm for the Vertex Feedback Set Problem Authors: Ann Becker and Dan.
CSL758 Instructors: Naveen Garg Kavitha Telikepalli Scribe: Manish Singh Vaibhav Rastogi February 7 & 11, 2008.
1 Discrete Structures & Algorithms Graphs and Trees: II EECE 320.
June 3, 2015Windows Scheduling Problems for Broadcast System 1 Amotz Bar-Noy, and Richard E. Ladner Presented by Qiaosheng Shi.
Graph Clustering. Why graph clustering is useful? Distance matrices are graphs  as useful as any other clustering Identification of communities in social.
Approximation Algorithms
Fast Distributed Algorithm for Convergecast in Ad Hoc Geometric Radio Networks Alex Kesselman, Darek Kowalski MPI Informatik.
Approximation Algorithms
Wroclaw University, Sept 18, Approximation via Doubling (Part II) Marek Chrobak University of California, Riverside Joint work with Claire Kenyon-Mathieu.
Krakow, Jan. 9, Outline: 1. Online bidding 2. Cow-path 3. Incremental medians (size approximation) 4. Incremental medians (cost approximation) 5.
Optimization problems INSTANCE FEASIBLE SOLUTIONS COST.
Computability and Complexity 24-1 Computability and Complexity Andrei Bulatov Approximation.
Preference Analysis Joachim Giesen and Eva Schuberth May 24, 2006.
Lecture 6: Point Location Computational Geometry Prof. Dr. Th. Ottmann 1 Point Location 1.Trapezoidal decomposition. 2.A search structure. 3.Randomized,
Performance guarantees for hierarchical clustering Sanjoy Dasgupta University of California, San Diego Philip Long Genomics Institute of Singapore.
A General Approach to Online Network Optimization Problems Seffi Naor Computer Science Dept. Technion Haifa, Israel Joint work: Noga Alon, Yossi Azar,
Priority Models Sashka Davis University of California, San Diego June 1, 2003.
1 Joint work with Shmuel Safra. 2 Motivation 3 Motivation.
Chapter 3: Cluster Analysis  3.1 Basic Concepts of Clustering  3.2 Partitioning Methods  3.3 Hierarchical Methods The Principle Agglomerative.
Randomized Algorithms - Treaps
Algorithms for Enumerating All Spanning Trees of Undirected and Weighted Graphs Presented by R 李孟哲 R 陳翰霖 R 張仕明 Sanjiv Kapoor and.
1 Introduction to Approximation Algorithms. 2 NP-completeness Do your best then.
The Selection Problem. 2 Median and Order Statistics In this section, we will study algorithms for finding the i th smallest element in a set of n elements.
Competitive Queue Policies for Differentiated Services Seminar in Packet Networks1 Competitive Queue Policies for Differentiated Services William.
Clustering.
Approximate Inference: Decomposition Methods with Applications to Computer Vision Kyomin Jung ( KAIST ) Joint work with Pushmeet Kohli (Microsoft Research)
BIRCH: Balanced Iterative Reducing and Clustering Using Hierarchies A hierarchical clustering method. It introduces two concepts : Clustering feature Clustering.
Linear Program Set Cover. Given a universe U of n elements, a collection of subsets of U, S = {S 1,…, S k }, and a cost function c: S → Q +. Find a minimum.
A Optimal On-line Algorithm for k Servers on Trees Author : Marek Chrobak Lawrence L. Larmore 報告人:羅正偉.
Database Management Systems, R. Ramakrishnan 1 Algorithms for clustering large datasets in arbitrary metric spaces.
The full Steiner tree problem Theoretical Computer Science 306 (2003) C. L. Lu, C. Y. Tang, R. C. T. Lee Reporter: Cheng-Chung Li 2004/06/28.
Heaps © 2010 Goodrich, Tamassia. Heaps2 Priority Queue ADT  A priority queue (PQ) stores a collection of entries  Typically, an entry is a.
1 Microarray Clustering. 2 Outline Microarrays Hierarchical Clustering K-Means Clustering Corrupted Cliques Problem CAST Clustering Algorithm.
NOTE: To change the image on this slide, select the picture and delete it. Then click the Pictures icon in the placeholder to insert your own image. Fast.
The geometric GMST problem with grid clustering Presented by 楊劭文, 游岳齊, 吳郁君, 林信仲, 萬高維 Department of Computer Science and Information Engineering, National.
Clustering Data Streams A presentation by George Toderici.
Approximation Algorithms based on linear programming.
Theory of Computational Complexity Probability and Computing Chapter Hikaru Inada Iwama and Ito lab M1.
Khaled M. Alzoubi, Peng-Jun Wan, Ophir Frieder
Heaps © 2010 Goodrich, Tamassia Heaps Heaps
Chapter 5. Optimal Matchings
Autumn 2016 Lecture 11 Minimum Spanning Trees (Part II)
Enumerating Distances Using Spanners of Bounded Degree
Autumn 2015 Lecture 11 Minimum Spanning Trees (Part II)
© 2013 Goodrich, Tamassia, Goldwasser
The Full Steiner tree problem Part Two
Compact routing schemes with improved stretch
BIRCH: Balanced Iterative Reducing and Clustering Using Hierarchies
Clustering.
Winter 2019 Lecture 11 Minimum Spanning Trees (Part II)
Hierarchical Clustering
Online Ranking for Tournament Graphs
Autumn 2019 Lecture 11 Minimum Spanning Trees (Part II)
Presentation transcript:

1 By: MOSES CHARIKAR, CHANDRA CHEKURI, TOMAS FEDER, AND RAJEEV MOTWANI Presented By: Sarah Hegab

2 Outline: Motivation Main Problem Hierarchical Agglomerative Clustering A Model Incremental Clustering Different incremental algorithms Lower Bounds for incremental algorithms Dual Problem

3 I. Main Problem The clustering problem is as follows: given n points in a metric space M, partition the points into k clusters so as to minimize the maximum cluster diameter.

4 1.Greedy Incremental Clustering a)Center-Greedy b)Diameter-greedy

5 a) Center-Greedy The center-greedy algorithm associates a center for each cluster and merges the two clusters whose centers are closest. The center of the old cluster with the larger radius becomes the new center Theorem: The center-greedy algorithm’s performance ratio has a lower bound of 2k - 1.

6 0 v1v1 v2v2 v3v3 v4v4 v5v S0S0 S1S1 S3S3 S2S2 S2S2 a) Center-Greedy cont. Proof: 1-Tree Construction K=2

7 a) Center-Greedy cont. 2-Tree  Graph Set A i (in our example A i ={{v 1 },{v 2 }, {v 3 },{v 4 }}) v1v1 v2v2 v3v3 v4v4 v5v5 S0S0 S1S1 S3S3 S2S2 S2S   1-   1-   1

8 Post-Order Traverse

9 a) Center-Greedy cont. Claims: For 1 <= i <= 2k - 1, A i is the set of clusters of center-greedy which contain more than one vertex after the k + i vertices v 1,..., v k+i are given. There is a k-clustering of G of diameter 1. The clustering which achieves the above diameter is {S 0 US 1,..., S 2k-2 US 2k-1 }.

10 K=4

11 Competitiveness of Center-Greedy Theorem : The center-greedy algorithm has performance ratio of 2k-1 in any metric space.

12 b) Diameter-Greedy The diameter-greedy algorithm always merges those two clusters which minimize the diameter of the resulting merged cluster. Theorem : The diameter-greedy algorithm’s performance ratio  (log(k)) is even on the line.

13 b) Diameter-Greedy cont. Proof: 1) Assumptions U i = U j=1 Fi {{pij, qij}, {rij, sij}}, V i = U j=1 Fi {{qij}, {rij}}, W i = U j=1 Fi {{pij}, {qij, rij}}, X i = U j=1 Fi {{pij}, {qij, rij}, {sij}}, Y i = U j=1 Fi {{pij, qij, rij}, {sij}}, Z i = U j=1 Fi {{pij, qij, rij, sij}}.

14 b) Diameter-Greedy cont. Proof: 2) Invariant : When the last element of K t is received, diameter-greedy’s k+1 clusters are (U i=1 t-2 Z i ) UY t-1 U X t (U r i=t+1 V i ). Since there are k+1 clusters, two of the clusters have to be merged and the algorithm merges two clusters in V t+1 to form a cluster of diameter (t+1). Without loss of generality, we may assume that the clusters merged are {q(t+1)1} and {r(t+1)1}.

15 Competitiveness of Diameter-Greedy Theorem : For k = 2, the diameter-greedy algorithm has a performance ratio 3 in any metric space.

16 2.Doubling Algorithm a)Deterministic b)Randomized c)Oblivious d)Randomized Oblivious

17 a) Deterministic doubling algorithm The algorithm works in phases At the start of phase i it has k+1 clusters Uses  and , s.t  /(1-  )<=  At start of phase i the following is assumed: 1. for each cluster C j, the radius of C j defined as max p  Cj d(c j, p) is at most αd i 2. for each pair of clusters C j and C l, the inter- center distance d(c j, c l ) => d i 3. d i <= opt.

18 a) Deterministic doubling algorithm Each phase has two stages 1- Merging stage, in which the algorithm reduces the number of clusters by merging certain pairs 2-Update stage, in which the algorithm accepts new updates and tries to maintain at most k clusters without increasing the radius of the clusters or violating the invariants A phase ends when number of clusters exceeds k

19 a) Deterministic doubling algorithm Definition: The t-threshold graph on a set of points P = {p 1, p 2,..., p n } is the graph G=(P,E) such that (p i, p j ) in E if and only if d(p i, p j ) <= t. Merging stage defines d i+1 =  d i and a graph G d i+1 –threshold for centers c 1,..., c k+1. New clusters C’ 1 …C’ m. If m=k+1 this ends the phase i

20 a) Deterministic doubling algorithm Lemma The pairwise distance between cluster centers after the merging stage of phase i is at least d i+1. Lemma The radius of the clusters after the merging stage of phase i is at most d i+1 +αd i <=αd i+1 Update continues while the number of clusters is at most k. It is restricted by the radius bound αd i+1. Then phase i ends.

21 a) Deterministic doubling algorithm Initialization : the algorithm waits until k+1 points have arrived then enters phase 1, with each point as a center containing just itself. And d 1 set to the distance between the closest pair of points

22 a) Deterministic doubling algorithm Lemma The k + 1 clusters at the end of the ith phase satisfy the following conditions: 1. The radius of the clusters is at most αd i The pairwise distance between the cluster centers is at least d i d i+1 <= OPT, where OPT is the diameter of the optimal clustering for the current set of points. Theorem: The doubling algorithm has performance ratio 8 in any metric space.

23 a) Deterministic doubling algorithm Example to show the analysis is tight: k=>3. Input consists of k+3 points p 1 …p k+3 the points p 1 …p k+1 have distance 1, p k+2,p k+3 have distance 4 from the others, and 8 from each other.

24 b) Randomized doubling algorithm Choose a random variable r from [1/e,1] according to the probability density function 1/r The min pairwise distance of the first k+1 point is x. And d 1 =rx  =e,  =e/(e-1)

25 b) Randomized doubling algorithm Theorem : The randomized doubling algorithm has expected performance ratio 2e in any metric space. The same bound is also achieved for the radius measure.

26 c) Oblivious clustering algorithm Does not need to know k Assume we have un upper bound on the max distance between point which is 1. Points are maintained in a tree

27 c) Oblivious clustering algorithm cont. At distance greater than 1/2 i Within dist. 1/2 i-1 from parent Where i is the depth of the vertex, i=>0 Root at depth 0

28 Illustration Of Oblivious clustering algorithm:

29 c) Oblivious clustering algorithm cont. How do we obtain the k clusters from the tree? If k is given, and i is the greatest depth containing at most k points. These are the k cluster centers. The sub-trees of the vertices at depth i are the clusters. As points are added, the number of vertices at depth i increases; if it goes beyond k, then we change i to i - 1, collapsing certain clusters; otherwise, the new point is inserted in one of the existing clusters.

30 c) Oblivious clustering algorithm cont. Theorem : The algorithm that outputs the k clusters obtained from the tree construction has performance ratio 8 for the diameter measure and the radius measure. Optimal diameter is ½ i+1 < d <= ½ I Then points at depth i are in different clusters, so there are at most k of them. j=>i be the greatest depth containing at most k points. Subtrees are at a distance of the root within ½ j + ½ j+1 + ½ j+2 + · · ·<= ½ j-1 < 4d.

31 d) Randomized Oblivious Distance threshold for depth i is r/e i r chosen once at random from [1,e], according to the PDF 1/r The expected diameter is at most 2e.OPT diameter

32 Lower Bounds Theorem1: For k = 2, there is a lower bound of 2 and 2 - ½ k/2 on the performance ratio of deterministic and randomized algorithms, respectively, for incremental clustering on the line.

33 Lower Bounds cont. Theorem2: There is a lower bound of 1+2 1/2 on the performance ratio of any deterministic incremental clustering algorithm for arbitrary metric spaces.

34 Lower Bounds cont.

35 Lower Bounds cont. Theorem3: For any e>0 and k=>2, there is a lower bound of 2 - e on the performance ratio of any randomized incremental algorithm.

36 Lower Bounds cont. Theorem4: For the radius measure, no deterministic incremental clustering algorithm has a performance ratio better than 3 and no randomized algorithm has a ratio better than 3 – e for any fixed e > 0.

37 II. Dual Problem For a sequence of points p 1,p 2,...,p n  R d, cover each point with a unit ball in d as it arrives, so as to minimize the total number of balls used.

38 II. Dual Problem Rogers Theorem: R d can be covered by any convex shape with covering density O(d log d). Theorem: For the dual clustering problem in R d, there is an incremental algorithm with performance ratio O(2 d d log d). Theorem: For the dual clustering problem in d, any incremental algorithm must have performance ratio  ( (log d)/(log log log d) ).

39