Small-world networks

Size: px
Start display at page:

Download "Small-world networks"

Transcription

1 Small-world networks c A. J. Ganesh, University of Bristol, 2015 Popular folklore asserts that any two people in the world are linked through a chain of no more than six mutual acquaintances, as encapsulated in the phrase six degrees of separation. Some evidence in support of this claim arose from the experiments of an American sociologist, Stanley Milgram, in the 1960s. He gave letters to people in Nebraska and Kansas, addressed to a recipient in Boston about whom they were provided with some information. They had to ensure that the letter reached the recipient, but only by forwarding it to someone they knew personally, who would have to forward it likewise until it reached the target. Each intermediary added their contact information to the letter, so that Milgram could reconstruct the chain it had passed through on route to the target. In these experiments, the median length of chain among those letters which reached the target (which were a minority of the set of letters initially handed out) was around six. Milgram s experiments have been subsequently repeated on and Instant messenger networks, with qualitatively similar results. This leads to the question of whether we can develop mathematical models of networks that can reflect these features of real-world networks of friendship or acquaintance. It turns out that very simply mathematical models called random graphs can indeed exhibit very small diameters. In this model, the number of nodes and the mean node degree (or, possibly, the degree distribution) are fixed, but the edges are then assigned randomly between nodes. (We won t describe the model of randomness in detail here as that is not the primary focus of this chapter, but you can look up the Erdős- Renyi random graph, and the configuration model for further details of such random graph models and their properties.) But it turns out that random graphs are not good models of social networks because they show no social structure - they lack triangles and other small dense subnetworks, which are typical of real social networks. This led Strogatz and Watts to propose a small network which is constructed by a two-stage process. In the first stage, we start with a base network which has considerably locality of 1

2 connection; an example might be nodes placed on a ring or a lattice, with each node connected to its k nearest neighbours, for some fixed parameter k that is not too large. In the second stage, some of these nearest neighbour links are selected and rewired randomly, meaning that the link is broken, and re-made with some other node chosen at random, possibly far away in the base network. Strogatz and Watts then look at the relationship between the number (or proportion) of such rewired links and the diameter of the resulting graph. It turns out that once a sufficiently high proportion of links is rewired, the diameter becomes small (the rewired links have a random graph structure), but the links that haven t been rewired retain the locality that is more typical of a social network. Thus, a fairly simple mathematical model can exhibit the feature of small diameter observed in real-world networks. This feature has come to be known as the small-world property, and networks of the form described above have come to be called small-world networks. Jon Kleinberg came to Milgram s experiments from a slightly different perspective. He observed that not only did the network of social acquaintance have small diameter, it was also navigable. What the letter term means is that many of the recipients of Milgram s letters were able to find short paths to the target addressee. This would be straightforward if they knew the global structure of the network (the acquaintance relationships among all people in the world), but obviously they cannot know this! How then did they find short paths to the target? One could conjecture plausible explanations in terms of the information they might have used; perhaps they tried to move geographically closer to the target, while also matching some other feature such as type of employment, or education institution attended, or church membership. But Kleinberg wanted to construct mathematical network models that would have the feature of being navigable, and understand what structural properties give rise to this feature. We shall describe his work in this chapter. Kleinberg considered the following, quite stylised, network model, which is very similar to the small-world model of Strogatz and Watts. We start with a finite d-dimenionsal lattice on n d points. For concreteness, we shall work with d = 2, though all the results generalise to arbitary d (and were proved in that setting by Kleinberg). Thus, we have a lattice on n 2 points. Instead of breaking and re-wiring some of these links, we instead simply add some more random links. This has the advantage that the lattice maintains connectivity (random rewiring could break the network into disjoint pieces), 2

3 and we only need to think about the diameter of the resulting network. The extra links are added as follows. Each node in the lattice has up to four nearest neighbour links (four in the interior, three on the boundary and two at the corners). In addition, each node chooses one other node at random, not necessarily uniformly, and establishes a link to it. A node x chooses a node y with probability proportional to r(x, y) s, where r(x, y) is the lattice distance between x and y (r(x, y) = x y 1 = x 1 y 1 + x 2 y 2, where x 1 and x 2 denote the first and second co-ordinate of node x and similarly for y), and s 0 is a parameter. The node y chosen according to this probability distribution is called the long-range contact of x, and we say that x establishes a shortcut to y. In addition, other nodes may also establish shortcuts to x. The choices of long-range contacts by distinct nodes are mutually independent. While the establishing of shortcuts is most easily described in directed terms, the link should be thought of as bi-directional, and the whole graph as undirected. The role of the parameter s is that it captures the characteristic length of the shortcuts. If s is large, then choosing y for which r(x, y) is large is highly unlikely; long links are penalised, and shortcuts typically go to nodes which are fairly close by. At the other extreme, if s = 0, then there is no dependence on distance, and all nodes are equally likely to be chosen as the long-range contact of x. Hence, the mean length of a shortcut will be of order n in this case, whereas, for sufficiently large s, it will be of order 1. We shall henceforth refer to the above graph model as Kleinberg s model, noting that it is parametrised by a single parameter s, which is a non-negative integer. Kleinberg wanted to know for what values of s the resulting graph is navigable. He established the remarkable result that there is only one value of s for which it is navigable, namely s = 2! (In the more general case of d dimensions, it is only navigable if s = d.) Before formally stating and proving this result, we need a mathematically precise defintion of navigability. But first, it is worth mentioning that navigability is a different property from having a small diameter, which is what we first considered in the context of small-world networks. It turns out that the above models have small diameter for all s sufficiently small, and certainly s 2, but are not navigable for s < 2. In order to define navigability, we first need to introduce the concept of a decentralised routing algorithm. An algorithm for routing a message from a source node u to a destination node v will forward this message through 3

4 a series of intermediate nodes: in other words, it will come up with a route u = x 0, x 1, x 2,..., x k = v along which the messages passes from u to v. In this case, we say that the message is delivered in k steps. The route has to be a path in the network, i.e., (x i, x i+1 ) has to be an edge, either an original lattice edge or a shortcut. But subject to this constraint, how is the route determined? The distinction between centralised and decentralised algorithms has to do with the information available to the routing algorithm at any stage. In both cases, we assume that the co-ordinates of the source, destination and current node are known. In addition, a centralised algorithm knows the full network (i.e., all the shortcuts) and can use this information to find an optimal route. There are a number of well-known algorithms for finding the shortest path between a source and a destination in a given graph; look up Dijkstra s algorithm, for example. On the other hand, we will call an algorithm decentralised if, at some intermediate stage after visiting nodes x 0, x 1,..., x m, the only information available to the algorithm is the coordinates of these nodes, and of their lattice and long-range neighbours. Thus, it has to choose the next node to visit based on this information only. However, we impose no constraints on how it should use this information; it may make arbitrarily complicated decisions, and these may be deterministic or random functions of the available information up to time m. Roughly speaking, we will prove the following results. We will show that, if s = 2, then a specific decentralised algorithm, namely the greedy algorithm (defined below) can efficiently route a message from an arbitrary source to an arbitrary destination. On the other hand, if s 2, then there is no decentralised algorithm that can efficiently route messages between arbitrary source-destination pairs. The notion of efficiency we use is the following: a message delivery algorithm on the n n grid is efficient if, for any sourcedestination pair, the expected number of messages required to deliver it is bounded above by a polynomial in the logarithm of n. If it is not so bounded, then it is inefficient. The expectation is with respect to the randomness in the distribution of shortcuts, as well as in the routing decisions of the algorithm if these are random. We will prove an explicit upper bound on the message delivery time of the greedy algorithm when s = 2, and an explicit lower bound for any decentralised algorithm when s 2. In the remainder of this chapter, distance will always refer to lattice distance, i.e., r(x, y) = x y 1. 4

5 Theorem 1 Consider Kleinberg s model with s = 2. Fix a source node u and a target node v, and consider the greedy routing algorithm where, at each step, the message is routed to whichever neighbour of the current node (lattice neighbour or long-range contact) is closest to the target; if there is more than one node at the same distance, the tie is broken uniformly at random. The expected number of steps that it takes for the message to reach its target is bounded above by c log 2 n, where c > 0 is a constant that does not depend on n, or on the choice of source and target nodes. Proof. We first outline the main steps in the proof. The proof will proceed by considering how long it takes before the distance to the target is halved. We will show that, at any step of the routing algorithm, there is at least a c/ log n chance that there is a shortcut from the current node that takes the message more than half-way towards the target. Hence, the number of steps before such a halving is stochastically dominated by a geometric random variable with mean c log n. We also note that, irrespective of where u and v are located, log 2 n = c log n halvings of the distance suffice to reach the target. Combining these yields the claim of the theorem. Throughout, we shall use c to denote a generic constant, not necessarily the same each time it is used. First, suppose that the message is currently at a node x, which is at distance D from the target v. Let U denote the set of all nodes within distance D/2 of the target. Let y be a node in U. Since r(x, v) = D and r(v, y) D/2, it follows from the triangle inequality that r(x, y) r(x, v) + r(v, y) 3D/2. This upper bound on the distance between x and y translates into a lower bound on the probability that x chooses y as its long-range contact, an event we denote by x y. We have P(x y) = r(x, y) 2 (3D/2) 2 r(x, z) 2 z x. (1) z x r(x, z) 2 The first equality is based on the model description, which says that P(x y) is proportional to r(x, y) 2 as s = 2; the denominator is the normalisation constant needed to make this a probability. We will now derive an upper bound on the denominator, in order to get a simpler lower bound on P(x y). The denominator involves a sum of r(x, z) 2 over all nodes z in the finite grid other than the node x itself. Let us rewrite this sum by grouping together all nodes which are at the same distance from x. It is fairly easy to see 5

6 that, for any d 1, the number of nodes at distance exactly d from x is at most 4d; in fact, it is exactly 4d in the infinite two-dimensional grid. Each of these nodes contributes a term d 2 to the sum. Moreover, the maximum possible distance between any two nodes in the finite grid is 2n, so it suffices to restrict the sum to d between 1 and 2n. Thus, we get r(x, z) 2 z x 2n d=1 4d d 2 4 ( 1 + 2n 0 x 1 dx ) = 4(1 + log(2n)). Substituting this in (1), we obtain that for any node y U, P(x y) 9 4D 2 (1 + log(2n)) c D 2 log n, (2) where c > 0 is some constant that does not depend on n. This lower bound holds for every node y U. Now, the node x chooses a single other node as its long-range contact; hence, for any two distinct nodes y and z, the events that x chooses y and that x chooses z are disjoint. Consequently, the probability that x chooses its long-range contact within the set U, which event we denote x U, is bounded below by P(x U) = y U P(x y) c D 2 U. (3) log n We now want a lower bound on U, the number of nodes that are within distance D/2 of the target node, v. First, we note that there are at least d nodes at distance exactly d, for any d n/2; this is because, if d n/2, then considering the square made up of 4d nodes at distance exactly d from v in the infinite two-dimensional lattice, at least one side of this square is completely within the finite grid made of n 2 points. (Drawing a picture might help you see this.) Hence, the number of nodes in U is bounded from below by U min(d/2,n/2) d=1 d min(d2, n 2 ) 8 D2 32, where the last inequality holds for all D 2n, which is the largest value of D we shall consider. Substituting this in (3), we get P(x U) c log n for some constant c that does not depend on n. (4) 6

7 On the event that {x U}, the greedy algorithm gets at least half-way to the target in the next routing step. Observe that the distance to the target decreases at each step under the greedy algorithm, because there is always a lattice neighbour of the current node which is one step closer to the target; hence, no node is visited twice by this algoithm. Since the longrange contacts of different nodes are chosen mutually independently, the probability of the event {x U}, conditional on the set of nodes visited so far, is the same as the unconditional probability. In other words, the lower bound in equation (4) holds for each node x, irrespective of history. This implies that the number of steps of the routing algorithm needed for this event occurs is stochastically dominated by a geometric random variable with success probability c/ log n. Consequently, the expected number of steps before the event occurs is bounded above by c log n for some (different) constant c. After this many steps, the algorithm halves the distance to the target. The rest of the proof is straightforward. Since log 2 n = log n/ log 2 = c log n halvings suffice to reach the target, and the expected number of steps between successive halvings is no more than c log n, the expected number of steps for the algorithm to deliver the message to the target is bounded above by c log 2 n, as claimed. Next, we consider the case s > 2 and show that, in this case, there is no decentralised algorithm that can achieve message delivery in a number of steps smaller than a (fractional) power of n, the number of nodes. In particular, no such algorithm can achieve message delivery in a polylogarithmic number of steps, which is our notion of efficiency. Theorem 2 Consider Kleinberg s model with s > 2. Fix a source node u and a target node v with r(u, v) > n/4. Then, for any decentralised routing algorithm, the expected number of steps for message delivery is at least cn (s 2)/(s 1), where c > 0 is a constant that does not depend on n, or on the choice of source and target nodes. The expectation is with respect to the random distribution of shortcuts, as well as any randomness in the routing algorithm. Remark. If the source and target are very close to each other, it may be possible for a routing algorithm to deliver the message in a small number of steps. The assumption that the distance between them is at least n/4 is not very stringent; it holds, for example, if the source and target are chosen 7

8 uniformly at random from among all nodes. The results of the theorem hold if the source-target separation is bigger than ɛn, for some ɛ > 0; the constant c in the statement of the theorem will then depend on ɛ as well. Proof. The outline of the proof is as follows. We shall show that the probability of encountering a long shortcut (where long will be precisely defined) at any given node is small. A decentralised algorithm has to make choices based on the nodes it has observed so far. It doesn t know if there is some other node nearby with a good shortcut. We will use an estimate of the probability of encountering a long shortcut, and the union bound, to show that the probability of encountering a long shortcut within the first cn (s 2)/(s 1) steps is smaller than a half. We will also show that, conditional on the event of not encountering long shortcuts, the routing algorithm could not have reached the target within this many steps. We now flesh out the details. First observe that for any two nodes x and y, the probability of x choosing y as its long-range contact is given by P(x y) = r(x, y) s z x r(x, z) s r(x, y) s, (5) since there is at least one node z at distance 1 from x, and so the denominator is bigger than 1. Now, fix a node x and let y be the random long-range contact chosen by it. Then, the probability that the distance between them is larger than k is bounded as follows: P(r(x, y) > k) = 2n j=k+1 z:r(x,z)=j) P(x z) 2n j=k+1 4j j s, (6) because there are at most 4j nodes at distance exactly j from x, and the probability of a shortcut to each of them is bounded above by j s by (5). Now, 2n j=k+1 4j j s 4 j=k+1 Substituting in (6), we find that j 1 s 4 P(r(x, y) > k) 4 s 2 k (s 2). k x 1 s dx = 4k2 s s 2. Exactly the same upper bound holds for the probability that some node y at distance bigger than k chooses x as its long-range contact. As the proof of 8

9 this follows along the same lines, we don t repeat it. Thus, the probability that any given node x has a long-range contact (that x chose, or that chose x) a distance further than k away is bounded by 8k (s 2) /(s 2). Let A x denote the event that node x has a shortcut of length bigger than n 1/(s 1). Then, as argued above, Next, let P(A x ) 8 s 2 s 2 n { s 2 n = min s 1. }n s 2 s 1, 16, 1 4 and, for a given decentralised routing algorithm, define A to be the event that one of the first n nodes visited by the routing algorithm has a shortcut of length bigger than n 1/(s 1). In other words, A = n i=1 A xi, where x i denotes the node visited at the i th step of the routing algorithm (or, more precisely, the i th distinct node visited by the algorithm, as it may backtrack). As the algorithm is decentralised, it has to choose nodes with no knowledge of their shortcuts. Hence, the bound on P(A x ) obtained above also applies to P(A xi ) for each i. Hence, by the union bound, P(A) P(A xi ) 8 s 2 s 2 n s 1 n 1 2. (7) n i=1 But on the event A c, the complement of the event A, no shortcuts of length bigger than n 1/(s 1) are encountered in the first n steps. Hence, by the triangle inequality, the total distance traversed in the first n steps is at most n n 1/(s 1). Substituting for n, we find that this distance is bounded above by 1 4 n s 2 1 n s 1 n s 1 = 4. But, by the assumption in the theorem, the source and target of the message are separated by a larger distance than this. Hence, on the event A c, the message is not delivered within the first n steps. Since the event A c has probability at least a half by (7), we conclude that the expected number of steps for message delivery is at least n /2. Looking at the expression for n, the theorem is proved. 9

10 Finally, we consider the case s < 2 and show that, in this case as well, there is no decentralised algorithm that can achieve message delivery in a number of steps smaller than a (fractional) power of n, the number of nodes. In particular, no such algorithm can achieve message delivery in a polylogarithmic number of steps. Theorem 3 Consider Kleinberg s model with s < 2. Fix a source node u and a target node v with r(u, v) > n/4. Then, for any decentralised routing algorithm, the expected number of steps for message delivery is at least cn (2 s)/3, where c > 0 is a constant that does not depend on n, or on the choice of source and target nodes. The expectation is with respect to the random distribution of shortcuts, as well as any randomness in the routing algorithm. Proof. The idea is as follows. We define a suitable neighbourhood U of the target node v and show that, for any node x, the probability of having a shortcut into U is small. This holds even for nodes x within U. Consequently, we argue that once this neighbourhood is reached, subsequent progress by the routing algorithm has to rely on lattice links because, with high probability, suitable shortcuts aren t available. We use this to obtain a lower bound on the expected number of routing steps. Intuitively, the key insight is that, when s < 2, shortcuts may have long range, but they are rather spread out. Hence, the routing algorithm may initially make rapid progress towards the target, but then gets stuck and is not able to zero in. Recall that for any two nodes x and y, the probability that x chooses y as its long-range contact is given by P(x y) = r(x, y) s z x r(x, 1, (8) z x z) s r(x, z) s since r(x, y) 1. We refine the upper bound on P(x y) by deriving a lower bound on the denominator. We have r(x, z) s j j s z x n/2 j=1 n/2 1 x 1 s dx (n/2)2 s 2 s. Why does the first inequality hold? In order to evaluate the sum, z x r(x, z) s, let us group together all nodes z at the same distance j from x. How many 10

11 such nodes are there? In the infinite two-dimensional lattice, there are exactly 4j of them, forming a square. If j n/2, then at least one side of this square is completely contained within the finite lattice being considered (just about, if x is at the centre of this lattice). Therefore, there are at least j nodes at distance exactly j from x, within the finite lattice. Each of these nodes contributes r(x, z) s = j s to the sum. Adding up the contributions from all nodes at distances between 1 and n/2, and ignoring contributions from nodes further away, gives us a lower bound on z x r(x, z) s and hence an upper bound on its reciprocal. Thus, substituting the above inequality in (8), we get P(x y) cn (2 s), (9) for some constant c. Now, define U = } {z : r(v, z) n 2 s 3, to be the set of all nodes wtihin distance n (2 s)/3 of the target v. Since the number of nodes at distance exactly j from v is at most 4j, the number of nodes in the set U is bounded by n (2 s)/3 U 1 + 4j cn 4 2s 3, j=1 for some constant c. (The extra 1 in U is for node v itself.) Let A x denote the event that there is a shortcut from x to U. For each node y U, the probability that x chooses y as its long-range contact is bounded above as in (9). Since the event of x choosing distinct nodes y in U as its contact are mutually exclusive events, the probability that x chooses some node in U as its contact is given by P(x U) = y U P(x y) cn (2 s) U cn 2 s 3, for some constant c. For each node y U, the probability that y chooses x as its long-range contact is exactly the same as the probability that x chooses y. Hence, a similar bound holds for the probability that some node in U chooses x as its long-range contact. Hence, we conclude that P(A x ) cn 2 s 3, (10) 11

12 for all nodes x, and for a constant c that does not depend on n or x. Let x i denote the i th distinct node visited by a given decentralised routing algorithm. Since the node choices of any such algorithm have to be made without knowledge of the shortcuts of the new node to be visited, the above bound on shortcut probability applies to each node x i. Define A = cn (2 s)/3 i=1 A xi, to be the event that there is no shortcut to the set U from any of the nodes visited in the first cn (2 s)/3 steps of the routing algorithm. Here c is a constant to be specified, not the same as the constant c in any other expression! Now, by (??) and the union bound, we can find a constant c such that P(A) cn (2 s)/3 i=1 P(A xi ) 1 2. (The constant 1/2 is arbitary above; we could have used any other constant, but the corresponding c would have been a different constant.) On the event A c, the complement of the event A, there is no shortcut into U in the first cn (2 s)/3 steps of the decentralised routing algorithm. Hence, even if the first step takes us to the boundary of the set U, further progress towards the target v has to be made using lattice steps. By the definition of the set U, at least n (2 s)/3 lattice steps are needed to reach v from the boundary of U. Since the event A c has probability bigger than a half, we conclude that the expected number of steps needed to reach the target is at least 1 2 min{c, 1}n(2 s)/3. This completes the proof of the theorem. 12

Math 443/543 Graph Theory Notes 10: Small world phenomenon and decentralized search

Math 443/543 Graph Theory Notes 10: Small world phenomenon and decentralized search Math 443/543 Graph Theory Notes 0: Small world phenomenon and decentralized search David Glickenstein November 0, 008 Small world phenomenon The small world phenomenon is the principle that all people

More information

Chapter 3. Set Theory. 3.1 What is a Set?

Chapter 3. Set Theory. 3.1 What is a Set? Chapter 3 Set Theory 3.1 What is a Set? A set is a well-defined collection of objects called elements or members of the set. Here, well-defined means accurately and unambiguously stated or described. Any

More information

1 Counting triangles and cliques

1 Counting triangles and cliques ITCSC-INC Winter School 2015 26 January 2014 notes by Andrej Bogdanov Today we will talk about randomness and some of the surprising roles it plays in the theory of computing and in coding theory. Let

More information

CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University

CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University http://cs224w.stanford.edu 10/4/2011 Jure Leskovec, Stanford CS224W: Social and Information Network Analysis, http://cs224w.stanford.edu

More information

Topic: Local Search: Max-Cut, Facility Location Date: 2/13/2007

Topic: Local Search: Max-Cut, Facility Location Date: 2/13/2007 CS880: Approximations Algorithms Scribe: Chi Man Liu Lecturer: Shuchi Chawla Topic: Local Search: Max-Cut, Facility Location Date: 2/3/2007 In previous lectures we saw how dynamic programming could be

More information

Notes for Lecture 24

Notes for Lecture 24 U.C. Berkeley CS170: Intro to CS Theory Handout N24 Professor Luca Trevisan December 4, 2001 Notes for Lecture 24 1 Some NP-complete Numerical Problems 1.1 Subset Sum The Subset Sum problem is defined

More information

2 The Fractional Chromatic Gap

2 The Fractional Chromatic Gap C 1 11 2 The Fractional Chromatic Gap As previously noted, for any finite graph. This result follows from the strong duality of linear programs. Since there is no such duality result for infinite linear

More information

ECE 158A - Data Networks

ECE 158A - Data Networks ECE 158A - Data Networks Homework 2 - due Tuesday Nov 5 in class Problem 1 - Clustering coefficient and diameter In this problem, we will compute the diameter and the clustering coefficient of a set of

More information

The Encoding Complexity of Network Coding

The Encoding Complexity of Network Coding The Encoding Complexity of Network Coding Michael Langberg Alexander Sprintson Jehoshua Bruck California Institute of Technology Email: mikel,spalex,bruck @caltech.edu Abstract In the multicast network

More information

How to explore big networks? Question: Perform a random walk on G. What is the average node degree among visited nodes, if avg degree in G is 200?

How to explore big networks? Question: Perform a random walk on G. What is the average node degree among visited nodes, if avg degree in G is 200? How to explore big networks? Question: Perform a random walk on G. What is the average node degree among visited nodes, if avg degree in G is 200? Questions from last time Avg. FB degree is 200 (suppose).

More information

/633 Introduction to Algorithms Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/27/18

/633 Introduction to Algorithms Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/27/18 601.433/633 Introduction to Algorithms Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/27/18 22.1 Introduction We spent the last two lectures proving that for certain problems, we can

More information

The strong chromatic number of a graph

The strong chromatic number of a graph The strong chromatic number of a graph Noga Alon Abstract It is shown that there is an absolute constant c with the following property: For any two graphs G 1 = (V, E 1 ) and G 2 = (V, E 2 ) on the same

More information

Theorem 2.9: nearest addition algorithm

Theorem 2.9: nearest addition algorithm There are severe limits on our ability to compute near-optimal tours It is NP-complete to decide whether a given undirected =(,)has a Hamiltonian cycle An approximation algorithm for the TSP can be used

More information

Treewidth and graph minors

Treewidth and graph minors Treewidth and graph minors Lectures 9 and 10, December 29, 2011, January 5, 2012 We shall touch upon the theory of Graph Minors by Robertson and Seymour. This theory gives a very general condition under

More information

M.E.J. Newman: Models of the Small World

M.E.J. Newman: Models of the Small World A Review Adaptive Informatics Research Centre Helsinki University of Technology November 7, 2007 Vocabulary N number of nodes of the graph l average distance between nodes D diameter of the graph d is

More information

Lecture 2 September 3

Lecture 2 September 3 EE 381V: Large Scale Optimization Fall 2012 Lecture 2 September 3 Lecturer: Caramanis & Sanghavi Scribe: Hongbo Si, Qiaoyang Ye 2.1 Overview of the last Lecture The focus of the last lecture was to give

More information

CSE 190 Lecture 16. Data Mining and Predictive Analytics. Small-world phenomena

CSE 190 Lecture 16. Data Mining and Predictive Analytics. Small-world phenomena CSE 190 Lecture 16 Data Mining and Predictive Analytics Small-world phenomena Another famous study Stanley Milgram wanted to test the (already popular) hypothesis that people in social networks are separated

More information

6th Bay Area Mathematical Olympiad

6th Bay Area Mathematical Olympiad 6th Bay Area Mathematical Olympiad February 4, 004 Problems and Solutions 1 A tiling of the plane with polygons consists of placing the polygons in the plane so that interiors of polygons do not overlap,

More information

Midterm 2 Solutions. CS70 Discrete Mathematics and Probability Theory, Spring 2009

Midterm 2 Solutions. CS70 Discrete Mathematics and Probability Theory, Spring 2009 CS70 Discrete Mathematics and Probability Theory, Spring 2009 Midterm 2 Solutions Note: These solutions are not necessarily model answers. Rather, they are designed to be tutorial in nature, and sometimes

More information

Overlay (and P2P) Networks

Overlay (and P2P) Networks Overlay (and P2P) Networks Part II Recap (Small World, Erdös Rényi model, Duncan Watts Model) Graph Properties Scale Free Networks Preferential Attachment Evolving Copying Navigation in Small World Samu

More information

The Probabilistic Method

The Probabilistic Method The Probabilistic Method Po-Shen Loh June 2010 1 Warm-up 1. (Russia 1996/4 In the Duma there are 1600 delegates, who have formed 16000 committees of 80 persons each. Prove that one can find two committees

More information

CPSC 536N: Randomized Algorithms Term 2. Lecture 5

CPSC 536N: Randomized Algorithms Term 2. Lecture 5 CPSC 536N: Randomized Algorithms 2011-12 Term 2 Prof. Nick Harvey Lecture 5 University of British Columbia In this lecture we continue to discuss applications of randomized algorithms in computer networking.

More information

TELCOM2125: Network Science and Analysis

TELCOM2125: Network Science and Analysis School of Information Sciences University of Pittsburgh TELCOM2125: Network Science and Analysis Konstantinos Pelechrinis Spring 2015 Figures are taken from: M.E.J. Newman, Networks: An Introduction 2

More information

Distributed Network Routing Algorithms Table for Small World Networks

Distributed Network Routing Algorithms Table for Small World Networks Distributed Network Routing Algorithms Table for Small World Networks Mudit Dholakia 1 1 Department of Computer Engineering, VVP Engineering College, Rajkot, 360005, India, Email:muditdholakia@gmail.com

More information

Network Mathematics - Why is it a Small World? Oskar Sandberg

Network Mathematics - Why is it a Small World? Oskar Sandberg Network Mathematics - Why is it a Small World? Oskar Sandberg 1 Networks Formally, a network is a collection of points and connections between them. 2 Networks Formally, a network is a collection of points

More information

A.1 Numbers, Sets and Arithmetic

A.1 Numbers, Sets and Arithmetic 522 APPENDIX A. MATHEMATICS FOUNDATIONS A.1 Numbers, Sets and Arithmetic Numbers started as a conceptual way to quantify count objects. Later, numbers were used to measure quantities that were extensive,

More information

CS264: Homework #4. Due by midnight on Wednesday, October 22, 2014

CS264: Homework #4. Due by midnight on Wednesday, October 22, 2014 CS264: Homework #4 Due by midnight on Wednesday, October 22, 2014 Instructions: (1) Form a group of 1-3 students. You should turn in only one write-up for your entire group. (2) Turn in your solutions

More information

6.207/14.15: Networks Lecture 5: Generalized Random Graphs and Small-World Model

6.207/14.15: Networks Lecture 5: Generalized Random Graphs and Small-World Model 6.207/14.15: Networks Lecture 5: Generalized Random Graphs and Small-World Model Daron Acemoglu and Asu Ozdaglar MIT September 23, 2009 1 Outline Generalized random graph models Graphs with prescribed

More information

Lecture 8: The Traveling Salesman Problem

Lecture 8: The Traveling Salesman Problem Lecture 8: The Traveling Salesman Problem Let G = (V, E) be an undirected graph. A Hamiltonian cycle of G is a cycle that visits every vertex v V exactly once. Instead of Hamiltonian cycle, we sometimes

More information

Topology Homework 3. Section Section 3.3. Samuel Otten

Topology Homework 3. Section Section 3.3. Samuel Otten Topology Homework 3 Section 3.1 - Section 3.3 Samuel Otten 3.1 (1) Proposition. The intersection of finitely many open sets is open and the union of finitely many closed sets is closed. Proof. Note that

More information

Clustering Using Graph Connectivity

Clustering Using Graph Connectivity Clustering Using Graph Connectivity Patrick Williams June 3, 010 1 Introduction It is often desirable to group elements of a set into disjoint subsets, based on the similarity between the elements in the

More information

ECS 253 / MAE 253, Lecture 8 April 21, Web search and decentralized search on small-world networks

ECS 253 / MAE 253, Lecture 8 April 21, Web search and decentralized search on small-world networks ECS 253 / MAE 253, Lecture 8 April 21, 2016 Web search and decentralized search on small-world networks Search for information Assume some resource of interest is stored at the vertices of a network: Web

More information

Calculus I Review Handout 1.3 Introduction to Calculus - Limits. by Kevin M. Chevalier

Calculus I Review Handout 1.3 Introduction to Calculus - Limits. by Kevin M. Chevalier Calculus I Review Handout 1.3 Introduction to Calculus - Limits by Kevin M. Chevalier We are now going to dive into Calculus I as we take a look at the it process. While precalculus covered more static

More information

Problem Set 1. Solution. CS4234: Optimization Algorithms. Solution Sketches

Problem Set 1. Solution. CS4234: Optimization Algorithms. Solution Sketches CS4234: Optimization Algorithms Sketches Problem Set 1 S-1. You are given a graph G = (V, E) with n nodes and m edges. (Perhaps the graph represents a telephone network.) Each edge is colored either blue

More information

Section 6.3: Further Rules for Counting Sets

Section 6.3: Further Rules for Counting Sets Section 6.3: Further Rules for Counting Sets Often when we are considering the probability of an event, that event is itself a union of other events. For example, suppose there is a horse race with three

More information

11.1 Facility Location

11.1 Facility Location CS787: Advanced Algorithms Scribe: Amanda Burton, Leah Kluegel Lecturer: Shuchi Chawla Topic: Facility Location ctd., Linear Programming Date: October 8, 2007 Today we conclude the discussion of local

More information

Switching for a Small World. Vilhelm Verendel. Master s Thesis in Complex Adaptive Systems

Switching for a Small World. Vilhelm Verendel. Master s Thesis in Complex Adaptive Systems Switching for a Small World Vilhelm Verendel Master s Thesis in Complex Adaptive Systems Division of Computing Science Department of Computer Science and Engineering Chalmers University of Technology Göteborg,

More information

Divisibility Rules and Their Explanations

Divisibility Rules and Their Explanations Divisibility Rules and Their Explanations Increase Your Number Sense These divisibility rules apply to determining the divisibility of a positive integer (1, 2, 3, ) by another positive integer or 0 (although

More information

MAE 298, Lecture 9 April 30, Web search and decentralized search on small-worlds

MAE 298, Lecture 9 April 30, Web search and decentralized search on small-worlds MAE 298, Lecture 9 April 30, 2007 Web search and decentralized search on small-worlds Search for information Assume some resource of interest is stored at the vertices of a network: Web pages Files in

More information

Lecture 15 : Review DRAFT

Lecture 15 : Review DRAFT CS/Math 240: Introduction to Discrete Mathematics 3/10/2011 Lecture 15 : Review Instructor: Dieter van Melkebeek Scribe: Dalibor Zelený DRAFT Today slectureservesasareviewofthematerialthatwillappearonyoursecondmidtermexam.

More information

Advanced Algorithms Class Notes for Monday, October 23, 2012 Min Ye, Mingfu Shao, and Bernard Moret

Advanced Algorithms Class Notes for Monday, October 23, 2012 Min Ye, Mingfu Shao, and Bernard Moret Advanced Algorithms Class Notes for Monday, October 23, 2012 Min Ye, Mingfu Shao, and Bernard Moret Greedy Algorithms (continued) The best known application where the greedy algorithm is optimal is surely

More information

Topology and Topological Spaces

Topology and Topological Spaces Topology and Topological Spaces Mathematical spaces such as vector spaces, normed vector spaces (Banach spaces), and metric spaces are generalizations of ideas that are familiar in R or in R n. For example,

More information

number Understand the equivalence between recurring decimals and fractions

number Understand the equivalence between recurring decimals and fractions number Understand the equivalence between recurring decimals and fractions Using and Applying Algebra Calculating Shape, Space and Measure Handling Data Use fractions or percentages to solve problems involving

More information

3 No-Wait Job Shops with Variable Processing Times

3 No-Wait Job Shops with Variable Processing Times 3 No-Wait Job Shops with Variable Processing Times In this chapter we assume that, on top of the classical no-wait job shop setting, we are given a set of processing times for each operation. We may select

More information

Lecture 5: Duality Theory

Lecture 5: Duality Theory Lecture 5: Duality Theory Rajat Mittal IIT Kanpur The objective of this lecture note will be to learn duality theory of linear programming. We are planning to answer following questions. What are hyperplane

More information

Lesson 18. Laura Ricci 08/05/2017

Lesson 18. Laura Ricci 08/05/2017 Lesson 18 WATTS STROGATZ AND KLEINBERG MODELS 08/05/2017 1 SMALL WORLD NETWORKS Many real networks are characterized by a diameter very low. In several social networks, individuals tend to group in clusters

More information

Planar Graphs. 1 Graphs and maps. 1.1 Planarity and duality

Planar Graphs. 1 Graphs and maps. 1.1 Planarity and duality Planar Graphs In the first half of this book, we consider mostly planar graphs and their geometric representations, mostly in the plane. We start with a survey of basic results on planar graphs. This chapter

More information

6.001 Notes: Section 4.1

6.001 Notes: Section 4.1 6.001 Notes: Section 4.1 Slide 4.1.1 In this lecture, we are going to take a careful look at the kinds of procedures we can build. We will first go back to look very carefully at the substitution model,

More information

CONNECTED SPACES AND HOW TO USE THEM

CONNECTED SPACES AND HOW TO USE THEM CONNECTED SPACES AND HOW TO USE THEM 1. How to prove X is connected Checking that a space X is NOT connected is typically easy: you just have to find two disjoint, non-empty subsets A and B in X, such

More information

Maximal Independent Set

Maximal Independent Set Chapter 0 Maximal Independent Set In this chapter we present a highlight of this course, a fast maximal independent set (MIS) algorithm. The algorithm is the first randomized algorithm that we study in

More information

Cuckoo Hashing for Undergraduates

Cuckoo Hashing for Undergraduates Cuckoo Hashing for Undergraduates Rasmus Pagh IT University of Copenhagen March 27, 2006 Abstract This lecture note presents and analyses two simple hashing algorithms: Hashing with Chaining, and Cuckoo

More information

2386 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 52, NO. 6, JUNE 2006

2386 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 52, NO. 6, JUNE 2006 2386 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 52, NO. 6, JUNE 2006 The Encoding Complexity of Network Coding Michael Langberg, Member, IEEE, Alexander Sprintson, Member, IEEE, and Jehoshua Bruck,

More information

Introduction to Algorithms / Algorithms I Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/18/14

Introduction to Algorithms / Algorithms I Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/18/14 600.363 Introduction to Algorithms / 600.463 Algorithms I Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/18/14 23.1 Introduction We spent last week proving that for certain problems,

More information

Order from Chaos. University of Nebraska-Lincoln Discrete Mathematics Seminar

Order from Chaos. University of Nebraska-Lincoln Discrete Mathematics Seminar Order from Chaos University of Nebraska-Lincoln Discrete Mathematics Seminar Austin Mohr Department of Mathematics Nebraska Wesleyan University February 8, 20 The (, )-Puzzle Start by drawing six dots

More information

Recitation 4: Elimination algorithm, reconstituted graph, triangulation

Recitation 4: Elimination algorithm, reconstituted graph, triangulation Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science 6.438 Algorithms For Inference Fall 2014 Recitation 4: Elimination algorithm, reconstituted graph, triangulation

More information

Lecture 3: Some Strange Properties of Fractal Curves

Lecture 3: Some Strange Properties of Fractal Curves Lecture 3: Some Strange Properties of Fractal Curves I have been a stranger in a strange land. Exodus 2:22 1. Fractal Strangeness Fractals have a look and feel that is very different from ordinary curves.

More information

Erdős-Rényi Model for network formation

Erdős-Rényi Model for network formation Network Science: Erdős-Rényi Model for network formation Ozalp Babaoglu Dipartimento di Informatica Scienza e Ingegneria Università di Bologna www.cs.unibo.it/babaoglu/ Why model? Simpler representation

More information

P Is Not Equal to NP. ScholarlyCommons. University of Pennsylvania. Jon Freeman University of Pennsylvania. October 1989

P Is Not Equal to NP. ScholarlyCommons. University of Pennsylvania. Jon Freeman University of Pennsylvania. October 1989 University of Pennsylvania ScholarlyCommons Technical Reports (CIS) Department of Computer & Information Science October 1989 P Is Not Equal to NP Jon Freeman University of Pennsylvania Follow this and

More information

Points covered an odd number of times by translates

Points covered an odd number of times by translates Points covered an odd number of times by translates Rom Pinchasi August 5, 0 Abstract Let T be a fixed triangle and consider an odd number of translated copies of T in the plane. We show that the set of

More information

EXTREME POINTS AND AFFINE EQUIVALENCE

EXTREME POINTS AND AFFINE EQUIVALENCE EXTREME POINTS AND AFFINE EQUIVALENCE The purpose of this note is to use the notions of extreme points and affine transformations which are studied in the file affine-convex.pdf to prove that certain standard

More information

function at the given point. Thus, for our previous example we would find that lim f(x) = 0

function at the given point. Thus, for our previous example we would find that lim f(x) = 0 Limits In order to introduce the notion of the it, we will consider the following situation. Suppose you are given a function, defined on some interval, except at a single point in the interval. What,

More information

Notes for Recitation 9

Notes for Recitation 9 6.042/18.062J Mathematics for Computer Science October 8, 2010 Tom Leighton and Marten van Dijk Notes for Recitation 9 1 Traveling Salesperson Problem Now we re going to talk about a famous optimization

More information

Trees. 3. (Minimally Connected) G is connected and deleting any of its edges gives rise to a disconnected graph.

Trees. 3. (Minimally Connected) G is connected and deleting any of its edges gives rise to a disconnected graph. Trees 1 Introduction Trees are very special kind of (undirected) graphs. Formally speaking, a tree is a connected graph that is acyclic. 1 This definition has some drawbacks: given a graph it is not trivial

More information

V1.0: Seth Gilbert, V1.1: Steven Halim August 30, Abstract. d(e), and we assume that the distance function is non-negative (i.e., d(x, y) 0).

V1.0: Seth Gilbert, V1.1: Steven Halim August 30, Abstract. d(e), and we assume that the distance function is non-negative (i.e., d(x, y) 0). CS4234: Optimisation Algorithms Lecture 4 TRAVELLING-SALESMAN-PROBLEM (4 variants) V1.0: Seth Gilbert, V1.1: Steven Halim August 30, 2016 Abstract The goal of the TRAVELLING-SALESMAN-PROBLEM is to find

More information

THREE LECTURES ON BASIC TOPOLOGY. 1. Basic notions.

THREE LECTURES ON BASIC TOPOLOGY. 1. Basic notions. THREE LECTURES ON BASIC TOPOLOGY PHILIP FOTH 1. Basic notions. Let X be a set. To make a topological space out of X, one must specify a collection T of subsets of X, which are said to be open subsets of

More information

round decimals to the nearest decimal place and order negative numbers in context

round decimals to the nearest decimal place and order negative numbers in context 6 Numbers and the number system understand and use proportionality use the equivalence of fractions, decimals and percentages to compare proportions use understanding of place value to multiply and divide

More information

Randomized Algorithms 2017A - Lecture 10 Metric Embeddings into Random Trees

Randomized Algorithms 2017A - Lecture 10 Metric Embeddings into Random Trees Randomized Algorithms 2017A - Lecture 10 Metric Embeddings into Random Trees Lior Kamma 1 Introduction Embeddings and Distortion An embedding of a metric space (X, d X ) into a metric space (Y, d Y ) is

More information

Generell Topologi. Richard Williamson. May 27, 2013

Generell Topologi. Richard Williamson. May 27, 2013 Generell Topologi Richard Williamson May 27, 2013 1 1 Tuesday 15th January 1.1 Topological spaces definition, terminology, finite examples Definition 1.1. A topological space is a pair (X, O) of a set

More information

Scribe from 2014/2015: Jessica Su, Hieu Pham Date: October 6, 2016 Editor: Jimmy Wu

Scribe from 2014/2015: Jessica Su, Hieu Pham Date: October 6, 2016 Editor: Jimmy Wu CS 267 Lecture 3 Shortest paths, graph diameter Scribe from 2014/2015: Jessica Su, Hieu Pham Date: October 6, 2016 Editor: Jimmy Wu Today we will talk about algorithms for finding shortest paths in a graph.

More information

Dynamic Programming Algorithms

Dynamic Programming Algorithms CSC 364S Notes University of Toronto, Fall 2003 Dynamic Programming Algorithms The setting is as follows. We wish to find a solution to a given problem which optimizes some quantity Q of interest; for

More information

Assignment 8; Due Friday, March 10

Assignment 8; Due Friday, March 10 Assignment 8; Due Friday, March 10 The previous two exercise sets covered lots of material. We ll end the course with two short assignments. This one asks you to visualize an important family of three

More information

Midterm 2 Solutions. CS70 Discrete Mathematics for Computer Science, Fall 2007

Midterm 2 Solutions. CS70 Discrete Mathematics for Computer Science, Fall 2007 CS70 Discrete Mathematics for Computer Science, Fall 007 Midterm Solutions Note: These solutions are not necessarily model answers Rather, they are designed to be tutorial in nature, and sometimes contain

More information

Notes on Minimum Spanning Trees. Red Rule: Given a cycle containing no red edges, select a maximum uncolored edge on the cycle, and color it red.

Notes on Minimum Spanning Trees. Red Rule: Given a cycle containing no red edges, select a maximum uncolored edge on the cycle, and color it red. COS 521 Fall 2009 Notes on Minimum Spanning Trees 1. The Generic Greedy Algorithm The generic greedy algorithm finds a minimum spanning tree (MST) by an edge-coloring process. Initially all edges are uncolored.

More information

2010 SMT Power Round

2010 SMT Power Round Definitions 2010 SMT Power Round A graph is a collection of points (vertices) connected by line segments (edges). In this test, all graphs will be simple any two vertices will be connected by at most one

More information

Lecture 4 Hierarchical clustering

Lecture 4 Hierarchical clustering CSE : Unsupervised learning Spring 00 Lecture Hierarchical clustering. Multiple levels of granularity So far we ve talked about the k-center, k-means, and k-medoid problems, all of which involve pre-specifying

More information

Combinatorial Gems. Po-Shen Loh. June 2009

Combinatorial Gems. Po-Shen Loh. June 2009 Combinatorial Gems Po-Shen Loh June 2009 Although this lecture does not contain many offical Olympiad problems, the arguments which are used are all common elements of Olympiad problem solving. Some of

More information

CHAPTER 8. Copyright Cengage Learning. All rights reserved.

CHAPTER 8. Copyright Cengage Learning. All rights reserved. CHAPTER 8 RELATIONS Copyright Cengage Learning. All rights reserved. SECTION 8.3 Equivalence Relations Copyright Cengage Learning. All rights reserved. The Relation Induced by a Partition 3 The Relation

More information

CHAPTER 2. Graphs. 1. Introduction to Graphs and Graph Isomorphism

CHAPTER 2. Graphs. 1. Introduction to Graphs and Graph Isomorphism CHAPTER 2 Graphs 1. Introduction to Graphs and Graph Isomorphism 1.1. The Graph Menagerie. Definition 1.1.1. A simple graph G = (V, E) consists of a set V of vertices and a set E of edges, represented

More information

MAS 341: GRAPH THEORY 2016 EXAM SOLUTIONS

MAS 341: GRAPH THEORY 2016 EXAM SOLUTIONS MS 41: PH THEOY 2016 EXM SOLUTIONS 1. Question 1 1.1. Explain why any alkane C n H 2n+2 is a tree. How many isomers does C 6 H 14 have? Draw the structure of the carbon atoms in each isomer. marks; marks

More information

Graph Theory Questions from Past Papers

Graph Theory Questions from Past Papers Graph Theory Questions from Past Papers Bilkent University, Laurence Barker, 19 October 2017 Do not forget to justify your answers in terms which could be understood by people who know the background theory

More information

3.7 Denotational Semantics

3.7 Denotational Semantics 3.7 Denotational Semantics Denotational semantics, also known as fixed-point semantics, associates to each programming language construct a well-defined and rigorously understood mathematical object. These

More information

Treaps. 1 Binary Search Trees (BSTs) CSE341T/CSE549T 11/05/2014. Lecture 19

Treaps. 1 Binary Search Trees (BSTs) CSE341T/CSE549T 11/05/2014. Lecture 19 CSE34T/CSE549T /05/04 Lecture 9 Treaps Binary Search Trees (BSTs) Search trees are tree-based data structures that can be used to store and search for items that satisfy a total order. There are many types

More information

γ(ɛ) (a, b) (a, d) (d, a) (a, b) (c, d) (d, d) (e, e) (e, a) (e, e) (a) Draw a picture of G.

γ(ɛ) (a, b) (a, d) (d, a) (a, b) (c, d) (d, d) (e, e) (e, a) (e, e) (a) Draw a picture of G. MAD 3105 Spring 2006 Solutions for Review for Test 2 1. Define a graph G with V (G) = {a, b, c, d, e}, E(G) = {r, s, t, u, v, w, x, y, z} and γ, the function defining the edges, is given by the table ɛ

More information

Order from Chaos. Nebraska Wesleyan University Mathematics Circle

Order from Chaos. Nebraska Wesleyan University Mathematics Circle Order from Chaos Nebraska Wesleyan University Mathematics Circle Austin Mohr Department of Mathematics Nebraska Wesleyan University February 2, 20 The (, )-Puzzle Start by drawing six dots at the corners

More information

arxiv: v1 [cs.ma] 8 May 2018

arxiv: v1 [cs.ma] 8 May 2018 Ordinal Approximation for Social Choice, Matching, and Facility Location Problems given Candidate Positions Elliot Anshelevich and Wennan Zhu arxiv:1805.03103v1 [cs.ma] 8 May 2018 May 9, 2018 Abstract

More information

Chapter 15 Introduction to Linear Programming

Chapter 15 Introduction to Linear Programming Chapter 15 Introduction to Linear Programming An Introduction to Optimization Spring, 2015 Wei-Ta Chu 1 Brief History of Linear Programming The goal of linear programming is to determine the values of

More information

9.5 Equivalence Relations

9.5 Equivalence Relations 9.5 Equivalence Relations You know from your early study of fractions that each fraction has many equivalent forms. For example, 2, 2 4, 3 6, 2, 3 6, 5 30,... are all different ways to represent the same

More information

Compact Sets. James K. Peterson. September 15, Department of Biological Sciences and Department of Mathematical Sciences Clemson University

Compact Sets. James K. Peterson. September 15, Department of Biological Sciences and Department of Mathematical Sciences Clemson University Compact Sets James K. Peterson Department of Biological Sciences and Department of Mathematical Sciences Clemson University September 15, 2017 Outline 1 Closed Sets 2 Compactness 3 Homework Closed Sets

More information

CS261: A Second Course in Algorithms Lecture #16: The Traveling Salesman Problem

CS261: A Second Course in Algorithms Lecture #16: The Traveling Salesman Problem CS61: A Second Course in Algorithms Lecture #16: The Traveling Salesman Problem Tim Roughgarden February 5, 016 1 The Traveling Salesman Problem (TSP) In this lecture we study a famous computational problem,

More information

Network Topology and Equilibrium Existence in Weighted Network Congestion Games

Network Topology and Equilibrium Existence in Weighted Network Congestion Games Network Topology and Equilibrium Existence in Weighted Network Congestion Games Igal Milchtaich, Bar-Ilan University August 2010 Abstract. Every finite noncooperative game can be presented as a weighted

More information

arxiv:submit/ [math.co] 9 May 2011

arxiv:submit/ [math.co] 9 May 2011 arxiv:submit/0243374 [math.co] 9 May 2011 Connectivity and tree structure in finite graphs J. Carmesin R. Diestel F. Hundertmark M. Stein 6 May, 2011 Abstract We prove that, for every integer k 0, every

More information

Highway Dimension and Provably Efficient Shortest Paths Algorithms

Highway Dimension and Provably Efficient Shortest Paths Algorithms Highway Dimension and Provably Efficient Shortest Paths Algorithms Andrew V. Goldberg Microsoft Research Silicon Valley www.research.microsoft.com/ goldberg/ Joint with Ittai Abraham, Amos Fiat, and Renato

More information

Mathematical and Algorithmic Foundations Linear Programming and Matchings

Mathematical and Algorithmic Foundations Linear Programming and Matchings Adavnced Algorithms Lectures Mathematical and Algorithmic Foundations Linear Programming and Matchings Paul G. Spirakis Department of Computer Science University of Patras and Liverpool Paul G. Spirakis

More information

A GRAPH FROM THE VIEWPOINT OF ALGEBRAIC TOPOLOGY

A GRAPH FROM THE VIEWPOINT OF ALGEBRAIC TOPOLOGY A GRAPH FROM THE VIEWPOINT OF ALGEBRAIC TOPOLOGY KARL L. STRATOS Abstract. The conventional method of describing a graph as a pair (V, E), where V and E repectively denote the sets of vertices and edges,

More information

14.1 Encoding for different models of computation

14.1 Encoding for different models of computation Lecture 14 Decidable languages In the previous lecture we discussed some examples of encoding schemes, through which various objects can be represented by strings over a given alphabet. We will begin this

More information

CMPSCI611: The Simplex Algorithm Lecture 24

CMPSCI611: The Simplex Algorithm Lecture 24 CMPSCI611: The Simplex Algorithm Lecture 24 Let s first review the general situation for linear programming problems. Our problem in standard form is to choose a vector x R n, such that x 0 and Ax = b,

More information

Towards the Proof of the PCP Theorem

Towards the Proof of the PCP Theorem CS640 Computational Complexity Towards the Proof of the PCP Theorem Instructor: Manindra Agrawal Scribe: Ramprasad Saptharishi Last class we saw completed our discussion on expander graphs. We shall now

More information

ELEC-270 Solutions to Assignment 5

ELEC-270 Solutions to Assignment 5 ELEC-270 Solutions to Assignment 5 1. How many positive integers less than 1000 (a) are divisible by 7? (b) are divisible by 7 but not by 11? (c) are divisible by both 7 and 11? (d) are divisible by 7

More information

Advanced Operations Research Prof. G. Srinivasan Department of Management Studies Indian Institute of Technology, Madras

Advanced Operations Research Prof. G. Srinivasan Department of Management Studies Indian Institute of Technology, Madras Advanced Operations Research Prof. G. Srinivasan Department of Management Studies Indian Institute of Technology, Madras Lecture 28 Chinese Postman Problem In this lecture we study the Chinese postman

More information

On ɛ-unit distance graphs

On ɛ-unit distance graphs On ɛ-unit distance graphs Geoffrey Exoo Department of Mathematics and Computer Science Indiana State University Terre Haute, IN 47809 g-exoo@indstate.edu April 9, 003 Abstract We consider a variation on

More information