Lecture 8: February 8

Size: px
Start display at page:

Download "Lecture 8: February 8"

Transcription

1 CS71 Randomness & Computation Spring 018 Instructor: Alistair Sinclair Lecture 8: February 8 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They may be distributed outside this class only with the permission of the Instructor. 8.1 The clique number of a random graph [cont.] We begin by sketching a proof of the following theorem from last lecture: Theorem 8.1 For G G n,p log 1/p (n. for any constant p (0, 1, the clique number of G is almost surely Proof: We restrict attention to the case p = 1/; the generalization to arbitrary p is straightforward. Throughout this proof, all logarithms are base. Let X k be the number of k-cliques in a graph G drawn from G n, 1. By linearity of expectation we have g(k := EX k = k ( k and let us define k0 (n to be the largest value of k such that g(k 1. An easy calculation (Exercise! shows that k 0 (n log n. (To see this is plausible, note that for k n, g(k is roughly nk k! k / k log n k / k log k. Now let c be some small integer constant (independent of n to be decided later. In order to prove our claim, it is enough to show the following, for some constant c: (1 for k 1 (n = k 0 (n + c : Pr[X k1(n > 0] 0 as n ( for k (n = k 0 (n c : Pr[X k(n > 0] 1 as n Note first that, if we work with expectations, then the above two claims are not hard to prove. Specifically, we can see that (a EX k1(n 0 as n (b EX k(n as n We will not rigorously prove these claims. However, they follow fairly easily by examining the rate of change of the function g(k near k = k 0. In particular, note that for k log n we have g(k + 1 g(k = n k k + 1 k n log n n 0 as n This indicates that, as n, in the neighborhood of k 0 (n the graph of g(k decreases sharply, so for some constant c, g(k 0 (n + c 0 as n and g(k 0 (n c as n. The details are left as an Exercise. We illustrate the behavior of g(k near k 0 by depicting log g(k in the neighborhood of k 0 (figure

2 8- Lecture 8: February 8 log g(k k k0 k1 k Figure 8.1: The descent of log g(k near k 0 (n for fixed n. Note that the domain of g(k is actually the integers, but we show an interpolated approximation for readability. Now, we can go ahead and prove claims (1 and (. Claim (1 is almost immediate: since X k is integer valued, we have Pr[X k1 > 0] = Pr[X k1 1], and then by Markov s inequality we have that Pr[X k1 1] EX k1. However, EX k1 0 as n and so, as n : Pr[X k1 > 0] 0 To prove Claim (, we will use the second moment method to show that Pr[X k = 0] 0 as n. Applying Chebyshev s inequality as we did for the case of 4-cliques earlier, we have Pr[X k = 0] Pr[ X k EX k EX k ] Var(X k (EX k. Thus, it is enough to show that Var(X k (EX k 0 as n. For ease of notation, let us denote the random variable X k by X and, for every subset S of the vertices of G of size k, let us denote by X S the following random variable: { 1, if S is a clique; X S = 0, if S is not a clique. Clearly, X = S X S. We now follow a similar path to our treatment of 4-cliques earlier, but in more generality and more systematically. First, we have Var(X = Var( S X S = S Var(X S + S T Cov(X S, X T. Writing S T to denote the fact that X S, X T are not independent, we have that Cov(X S, X T = 0 unless X S X T. In our current application, S T iff S T and S T. Continuing: Var(X = S S = S Var(X S + S T Cov(X S, X T E(X S + S T E(X S X T EX S + S T E(X S X T (since X S is 0/1 valued = EX + S T E(X S X T. It follows that Var(X (EX 1 EX + 1 (EX S T E(X SX T. Thus, since EX, it is enough for us to prove that S T E(X SX T = o((ex.

3 Lecture 8: February We can go further than this using symmetry of the events X S. 1 We have Pr[X S = 1 X T = 1] S T E(X S X T = S T = S T Pr[X S = 1]Pr[X T = 1 X S = 1] = ( Pr[X S = 1] Pr[X T = 1 X S = 1] S T :T S = Pr[X T = 1 X S0 = 1] Pr[X S = 1] (for any fixed S 0, by symmetry T :T S 0 S = EX Pr[X T = 1 X S0 = 1], T :T S 0 and so Var(X (EX 1 EX + 1 Pr[X T = 1 X S0 = 1]. EX T :T S 0 Thus, since EX, to conclude our proof it is enough to show that T :T S 0 Pr[X T = 1 X S0 = 1] = o(ex for a fixed S 0. Note that the above argument is very general and applies in any symmetric setting. Specializing now to our clique problem, and letting S 0 be any fixed set of vertices of size k (n, we have k 1 ( ( k n k Pr[X T = 1 X S0 = 1] = [(k ( ] i i k T :T S 0 i= i Note that the sum on the right hand side of the above equation is taken over i = T S 0. The factor ( k i counts the number of ways that T can choose i vertices of S 0 to share. Factor k k i is all the possible ways that T can choose its other vertices, whereas [(k ( ] i is the probability that all edges in T that are not shared with S 0 appear. Now we have the following: T :T S 0 Pr[X T = 1 X S0 = 1] = EX = = k 1 ( k i= k 1 i= k 1 i= k i k i k ( k (i k ( k k i k i [( k ( i ] f(i, where f(i := k max {f(i}. i k 1 ( k k i k i (i k Now it can be shown that, provided the constant c is chosen large enough, f(i is maximized for i = (Exercise!. Its value there is ( k k f( = k ( k k (as k n. k Since k log n, k 5 /n 0 as n, so the above quotient tends to zero as required. This concludes the proof of claim ( and the proof of the theorem. 1 i.e., there is an automorphism of the probability space that maps any X S onto any other n

4 8-4 Lecture 8: February 8 8. Random k-sat Background and Motivation Determining whether a k-cnf formula is satisfiable is an NP-complete problem. However, the problem may be easier to solve for random formulae. This is because of the empirical observation that a random formula with a certain density of clauses is either almost surely satisfiable or almost surely unsatisfiable depending on whether the density is below or above a certain critical value. There are ties between this threshold phenomenon and so-called spin glasses in statistical physics; indeed, some people believe that this connection may one day unlock the P vs NP question. Model We now define a model. ϕ k (n, rn is a k-cnf formula over n variables generated by choosing rn clauses independently and uniformly at random. What exactly do we mean by choosing the clauses randomly? The two standard possibilities are (i to choose the k literals in the clause independently and u.a.r. from the n possibilities (so that repetitions of variables are allowed; or (ii to choose the clause u.a.r. from the set of all k k valid k-clauses (i.e., without repeated variables. It turns out that most results hold for either model, so we will use the most convenient implementation in each application. The above model has just one parameter, r, which we can think of as the density of the formula. Obviously, as r increases there are more constraints so it becomes less likely that ϕ k (n, rn is satisfiable. Conjecture 8. For every k, there exists a threshold value rk R such that r > rk Pr[ϕ k (n, rn is satisfiable] 0 as n r < rk Pr[ϕ k (n, rn is satisfiable] 1 as n This conjecture has been a major open problem for at least 5 years. The first landmark result was the following, which resolves the case k =. Theorem 8.3 [CR9, FdlV9, G96] The threshold r exists and is equal to 1. Some years later, a weaker form of the conjecture was proved by Friedgut, who showed the existence of a non-uniform threshold for every k: Theorem 8.4 [F99] For every k, there exists a sequence {r k (n} such that ɛ > 0 r > (1 + ɛr k (n Pr[ϕ k (n, rn is satisfiable] 0 as n r < (1 ɛr k (n Pr[ϕ k (n, rn is satisfiable] 1 as n. The crucial difference between Theorem 8.4 and Conjecture 8. is that in the conjecture rk is a constant, while in the theorem r k (n may depend on n. Despite this weakness, Friedgut s result was a surprising and significant step. Following a long sequence of developments by several authors, a major breakthrough was achieved very recently in a technical tour-de-force by Ding, Sly and Sun: Theorem 8.5 [DSS14] For all sufficiently large k, the threshold r k exists.

5 Lecture 8: February The actual value of the threshold is also given in [DSS14] as the minimum of an explicit function. As had been previously established by Coja-Oghlan and Panagiotou [CP16], the asymptotic form of r k is where ε k 0 as k. r k = k ln 1 (1 + ln ε k, For small values of k >, the conjecture remains open. There has been much work establishing bounds on r k (assuming it exists for the important case of k = 3 (3-Sat. For example, it is known that 3.5 < r 3 < 4.49, and based on large-scale experiments and heuristic arguments, r 3 is believed to be close to 4.. The upper bound here is due to [DKMP09], and the lower bound to [KKL03,HS03]. Upper Bounds on rk. A more or less tight upper bound on r k can be found by a simple application of the first moment method. Let X = the number of satisfying assignments of ϕ k (r, rn. Then X = σ X σ where { 1, if assignment σ satisfies ϕ X σ = 0, if assignment σ does not satisfy ϕ Assuming the model in which every literal is independent, we clearly have E[X σ ] = (1 k rn, and therefore E[X] = n (1 k rn. Now it is easy to see that, if r k ln, E[X] 0 as n. By Markov s inequality, this immediately yields: Corollary 8.6 r k k ln. Lower Bounds on rk. A natural way to prove lower bounds is via algorithms: if a certain algorithm finds a satisfying assignment with high probability in some range of r, then obviously a satisfying assignment must exist in this range. However, bounds found in this way (see, e.g., [CF90] usually have the form rk c k k for some constant c, which is a factor O(k off from the upper bound. An important step was the introduction of non-algorithmic proofs based on the second moment method [AM0,AP04] to yield the much better bound r k k ln k for all k 3. This result already bounds the value of rk up to a linear term in k. The works [CP16,DSS14] mentioned earlier use more sophisticated non-algorithmic ideas inspired by statistical physics to obtain the final tight bound. The gap between algorithmic and non-algorithmic lower bounds suggests that there may exist two different thresholds: the first separating those densities for which it is easy to find a satisfying assignment w.h.p. (and therefore one necessarily exists and those for which a satisfying assignment exists w.h.p. but is hard to find; the second threshold separating the latter densities from those for which no satisfying assignment exists w.h.p. In the next lecture we will sketch the second moment approach to this lower bound. For now, we return to the proof of theorem 8.3. Proof of Theorem 8.3: We begin with the lower bound, namely that if r 1 ɛ then, with high probability, a random -SAT formula with m = rn (1 ɛn clauses is satisfiable. Let us recall the well-known fact (easy exercise that a -SAT formula is satisfiable if and only if its graph of implications does not contain a (directed cycle that contains both a variable and its negation. (The graph of implications contains a vertex for every literal and, for every clause (l 1 l, the edges ( l 1, l and ( l, l 1. For example, the edges induced by clause (x 1 x are as shown in Figure 8.. We will actually work with a slightly stronger sufficient condition which is easier to quantify, namely: the formula is satisfiable if it does not contain a bicycle, which is defined as follows.

6 8-6 Lecture 8: February 8 X1 X X1 X Figure 8.: The edges induced by clause (x 1 x. Definition 8.7 For k, a bicycle is a path (u, w 1, w,..., w k, v in the graph of implications, where the w i s are literals on distinct variables and u, v {w i, w i : 1 i k}. Exercise: Verify that if φ does not contain a bicycle then φ is satisfiable. Note that the variables appearing among the w i must be distinct! Therefore we may bound the probability that φ is not satisfiable as follows: Pr[ϕ not satisfiable] Pr[ϕ contains a bicycle] ( n n k k (k m k+1 Here we are summing over the possible sizes k of the bicycle; factor n k is an upper bound on the number of ways to choose k distinct variables for the bicycle, k accounts for the sign of each variable in the bicycle, (k is the number of ways to choose u and v, m k+1 is an upper bound on the number of ways to choose ( k+1 1 which of the m clauses of the formula induce the k + 1 edges of the bicycle, and is the probability 4 that the chosen clauses induce the corresponding edges of the bicycle. (Here we are working with the model which chooses valid clauses u.a.r. Now, the right hand side of the previous inequality becomes ( n n k k (k m k+1 k= 1 4 k+1 = n k= k= 1 4 n ( k+1 m k 0 as n n 1 The last expression tends to zero because the factor outside the summation goes to 0 and the summation is bounded by a constant. (This latter fact follows because m n 1 = (1 ɛn n 1 is less than 1, so the terms of the sum form a decreasing geometric series weighted by a polynomial factor. This concludes the proof of the first part (lower bound of theorem 8.3. The upper bound, which uses the second moment method, will be proved in the next lecture. k+1 References [AM0] [AP04] [CF90] D. Achlioptas and C. Moore, The asymptotic order of the random k-sat threshold, Proceedings of the 43rd IEEE FOCS, 00, pp D. Achlioptas and Y. Peres, The threshold for random k-sat is k log O(k, Journal of the American Mathematical Society 17 (004, pp M.-T. Chao and J. Franco, Probabilistic analysis of a generalization of the unit-clause literal selection heuristics for the k-satisfiability problem, Information Science 51 (1990, pp

7 Lecture 8: February [CR9] [CP16] [DKMP09] [DSS14] V. Chvátal and B. Reed, Mick gets some (the odds are on his side, Proceedings of the 33rd IEEE FOCS, 199, pp A. Coja-Oghlan and K. Panagiotou, The asymptotic k-sat threshold, Advances in Mathematics 88 (016, pp J. Diaz, L. Kirousis, D. Mitsche and X. Perez-Gimenez, On the satisfiability threshold of formulas with three literals per clause, Theoretical Computer Science 410 (009, pp J. Ding, A. Sly and N. Sun. Proof of the satisfiability conjecture for large k, ArXiv preprint arxiv: [math.pr], 014. [FdlV9] W. Fernandez de la Vega, On random -SAT, Manuscript, 199. [F99] E. Friedgut, Necessary and sufficient conditions for sharp thresholds of graph properties, Journal of the American Mathematical Society 1 (1999, pp [G96] A. Goerdt, A threshold for unsatisfiability, Journal of Computer & System Sciences 53 (1996, pp [HS03] [KKL03] M. Hajiaghayi and G.B. Sorkin, The satisfiability threshold for random 3-SAT is at least 3.5, submitted, 003. A. Kaporis, L.M. Kirousis and E. Lalas, Selecting complementary pairs of literals, Proc. LICS 03 Workshop on Typical Case Complexity and Phase Transitions, June 003.

Lecture 7: February 6

Lecture 7: February 6 CS271 Randomness & Computation Spring 2018 Instructor: Alistair Sinclair Lecture 7: February 6 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They

More information

Tight Bounds For Random MAX 2-SAT

Tight Bounds For Random MAX 2-SAT Tight Bounds For Random MAX 2-SAT MohammadTaghi Hajiaghayi Jeong Han Kim Abstract For a conjunctive normal form formula F with n variables and m = cn 2-variable clauses (c is called the density), denote

More information

Lecture 5: January 30

Lecture 5: January 30 CS71 Randomness & Computation Spring 018 Instructor: Alistair Sinclair Lecture 5: January 30 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They

More information

Lecture 28: April 26

Lecture 28: April 26 CS271 Randomness & Computation Spring 2018 Instructor: Alistair Sinclair Lecture 28: April 26 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They

More information

Two-coloring random hypergraphs

Two-coloring random hypergraphs Two-coloring random hypergraphs Dimitris Achlioptas Jeong Han Kim Michael Krivelevich Prasad Tetali December 17, 1999 Technical Report MSR-TR-99-99 Microsoft Research Microsoft Corporation One Microsoft

More information

Lecture 6: September 22

Lecture 6: September 22 CS294 Markov Chain Monte Carlo: Foundations & Applications Fall 2009 Lecture 6: September 22 Lecturer: Prof. Alistair Sinclair Scribes: Alistair Sinclair Disclaimer: These notes have not been subjected

More information

Notes 6 : First and second moment methods

Notes 6 : First and second moment methods Notes 6 : First and second moment methods Math 733-734: Theory of Probability Lecturer: Sebastien Roch References: [Roc, Sections 2.1-2.3]. Recall: THM 6.1 (Markov s inequality) Let X be a non-negative

More information

Phase Transitions and Satisfiability Threshold

Phase Transitions and Satisfiability Threshold Algorithms Seminar 2001 2002, F. Chyzak (ed.), INRIA, (200, pp. 167 172. Available online at the URL http://algo.inria.fr/seminars/. Phase Transitions and Satisfiability Threshold Olivier Dubois (a) and

More information

The Probabilistic Method

The Probabilistic Method The Probabilistic Method Janabel Xia and Tejas Gopalakrishna MIT PRIMES Reading Group, mentors Gwen McKinley and Jake Wellens December 7th, 2018 Janabel Xia and Tejas Gopalakrishna Probabilistic Method

More information

Independence and chromatic number (and random k-sat): Sparse Case. Dimitris Achlioptas Microsoft

Independence and chromatic number (and random k-sat): Sparse Case. Dimitris Achlioptas Microsoft Independence and chromatic number (and random k-sat): Sparse Case Dimitris Achlioptas Microsoft Random graphs W.h.p.: with probability that tends to 1 as n. Hamiltonian cycle Let τ 2 be the moment all

More information

Approximation Algorithm for Random MAX-kSAT

Approximation Algorithm for Random MAX-kSAT Approximation Algorithm for Random MAX-kSAT Yannet Interian Center for Applied Mathematics. Cornell University. Ithaca, NY 14853, USA interian@cam.cornell.edu Abstract. An approximation algorithm for random

More information

NP-Hardness reductions

NP-Hardness reductions NP-Hardness reductions Definition: P is the class of problems that can be solved in polynomial time, that is n c for a constant c Roughly, if a problem is in P then it's easy, and if it's not in P then

More information

Lecture 18: March 15

Lecture 18: March 15 CS71 Randomness & Computation Spring 018 Instructor: Alistair Sinclair Lecture 18: March 15 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They may

More information

Lecture 24: April 12

Lecture 24: April 12 CS271 Randomness & Computation Spring 2018 Instructor: Alistair Sinclair Lecture 24: April 12 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They

More information

Chasing the k-sat Threshold

Chasing the k-sat Threshold Chasing the k-sat Threshold Amin Coja-Oghlan Goethe University Frankfurt Random Discrete Structures In physics, phase transitions are studied by non-rigorous methods. New mathematical ideas are necessary

More information

Notes for Lecture 2. Statement of the PCP Theorem and Constraint Satisfaction

Notes for Lecture 2. Statement of the PCP Theorem and Constraint Satisfaction U.C. Berkeley Handout N2 CS294: PCP and Hardness of Approximation January 23, 2006 Professor Luca Trevisan Scribe: Luca Trevisan Notes for Lecture 2 These notes are based on my survey paper [5]. L.T. Statement

More information

c 2006 Society for Industrial and Applied Mathematics

c 2006 Society for Industrial and Applied Mathematics SIAM J. COMPUT. Vol. 36, No. 3, pp. 740 762 c 2006 Society for Industrial and Applied Mathematics RANDOM k-sat: TWO MOMENTS SUFFICE TO CROSS A SHARP THRESHOLD DIMITRIS ACHLIOPTAS AND CRISTOPHER MOORE Abstract.

More information

Essential facts about NP-completeness:

Essential facts about NP-completeness: CMPSCI611: NP Completeness Lecture 17 Essential facts about NP-completeness: Any NP-complete problem can be solved by a simple, but exponentially slow algorithm. We don t have polynomial-time solutions

More information

The unsatisfiability threshold conjecture: the techniques behind upper bound improvements

The unsatisfiability threshold conjecture: the techniques behind upper bound improvements The unsatisfiability threshold conjecture: the techniques behind upper bound improvements Lefteris M. Kirousis Yannis C. Stamatiou Michele Zito Abstract One of the most challenging problems in probability

More information

NP Completeness and Approximation Algorithms

NP Completeness and Approximation Algorithms Chapter 10 NP Completeness and Approximation Algorithms Let C() be a class of problems defined by some property. We are interested in characterizing the hardest problems in the class, so that if we can

More information

Homework 4 Solutions

Homework 4 Solutions CS 174: Combinatorics and Discrete Probability Fall 01 Homework 4 Solutions Problem 1. (Exercise 3.4 from MU 5 points) Recall the randomized algorithm discussed in class for finding the median of a set

More information

On the Solution-Space Geometry of Random Constraint Satisfaction Problems

On the Solution-Space Geometry of Random Constraint Satisfaction Problems On the Solution-Space Geometry of Random Constraint Satisfaction Problems Dimitris Achlioptas Department of Computer Science, University of California Santa Cruz optas@cs.ucsc.edu Amin Coja-Oghlan School

More information

The concentration of the chromatic number of random graphs

The concentration of the chromatic number of random graphs The concentration of the chromatic number of random graphs Noga Alon Michael Krivelevich Abstract We prove that for every constant δ > 0 the chromatic number of the random graph G(n, p) with p = n 1/2

More information

Hardness of Approximation

Hardness of Approximation Hardness of Approximation We have seen several methods to find approximation algorithms for NP-hard problems We have also seen a couple of examples where we could show lower bounds on the achievable approxmation

More information

CS5314 Randomized Algorithms. Lecture 18: Probabilistic Method (De-randomization, Sample-and-Modify)

CS5314 Randomized Algorithms. Lecture 18: Probabilistic Method (De-randomization, Sample-and-Modify) CS5314 Randomized Algorithms Lecture 18: Probabilistic Method (De-randomization, Sample-and-Modify) 1 Introduce two topics: De-randomize by conditional expectation provides a deterministic way to construct

More information

3-Coloring of Random Graphs

3-Coloring of Random Graphs 3-Coloring of Random Graphs Josep Díaz 1 UPC 1 Research partially supported by the Future and Emerging Technologies programme of the EU under contract 001907 (DELIS) Coloring problems. Given a graph G

More information

Stanford University CS254: Computational Complexity Handout 8 Luca Trevisan 4/21/2010

Stanford University CS254: Computational Complexity Handout 8 Luca Trevisan 4/21/2010 Stanford University CS254: Computational Complexity Handout 8 Luca Trevisan 4/2/200 Counting Problems Today we describe counting problems and the class #P that they define, and we show that every counting

More information

Aditya Bhaskara CS 5968/6968, Lecture 1: Introduction and Review 12 January 2016

Aditya Bhaskara CS 5968/6968, Lecture 1: Introduction and Review 12 January 2016 Lecture 1: Introduction and Review We begin with a short introduction to the course, and logistics. We then survey some basics about approximation algorithms and probability. We also introduce some of

More information

NP Completeness. CS 374: Algorithms & Models of Computation, Spring Lecture 23. November 19, 2015

NP Completeness. CS 374: Algorithms & Models of Computation, Spring Lecture 23. November 19, 2015 CS 374: Algorithms & Models of Computation, Spring 2015 NP Completeness Lecture 23 November 19, 2015 Chandra & Lenny (UIUC) CS374 1 Spring 2015 1 / 37 Part I NP-Completeness Chandra & Lenny (UIUC) CS374

More information

Random Satisfiability

Random Satisfiability Handbook of Satisfiability Armin Biere, Marijn Heule, Hans van Maaren and Toby Walsch IOS Press, 2008 c 2008 Dimitris Achlioptas. All rights reserved. 243 Chapter 8 Random Satisfiability Dimitris Achlioptas

More information

CS21 Decidability and Tractability

CS21 Decidability and Tractability CS21 Decidability and Tractability Lecture 20 February 23, 2018 February 23, 2018 CS21 Lecture 20 1 Outline the complexity class NP NP-complete probelems: Subset Sum NP-complete problems: NAE-3-SAT, max

More information

Lecture 3: September 10

Lecture 3: September 10 CS294 Markov Chain Monte Carlo: Foundations & Applications Fall 2009 Lecture 3: September 10 Lecturer: Prof. Alistair Sinclair Scribes: Andrew H. Chan, Piyush Srivastava Disclaimer: These notes have not

More information

Induced subgraphs with many repeated degrees

Induced subgraphs with many repeated degrees Induced subgraphs with many repeated degrees Yair Caro Raphael Yuster arxiv:1811.071v1 [math.co] 17 Nov 018 Abstract Erdős, Fajtlowicz and Staton asked for the least integer f(k such that every graph with

More information

Oberwolfach workshop on Combinatorics and Probability

Oberwolfach workshop on Combinatorics and Probability Apr 2009 Oberwolfach workshop on Combinatorics and Probability 1 Describes a sharp transition in the convergence of finite ergodic Markov chains to stationarity. Steady convergence it takes a while to

More information

Containment restrictions

Containment restrictions Containment restrictions Tibor Szabó Extremal Combinatorics, FU Berlin, WiSe 207 8 In this chapter we switch from studying constraints on the set operation intersection, to constraints on the set relation

More information

Lecture 04: Balls and Bins: Birthday Paradox. Birthday Paradox

Lecture 04: Balls and Bins: Birthday Paradox. Birthday Paradox Lecture 04: Balls and Bins: Overview In today s lecture we will start our study of balls-and-bins problems We shall consider a fundamental problem known as the Recall: Inequalities I Lemma Before we begin,

More information

1 More finite deterministic automata

1 More finite deterministic automata CS 125 Section #6 Finite automata October 18, 2016 1 More finite deterministic automata Exercise. Consider the following game with two players: Repeatedly flip a coin. On heads, player 1 gets a point.

More information

CS261: A Second Course in Algorithms Lecture #18: Five Essential Tools for the Analysis of Randomized Algorithms

CS261: A Second Course in Algorithms Lecture #18: Five Essential Tools for the Analysis of Randomized Algorithms CS261: A Second Course in Algorithms Lecture #18: Five Essential Tools for the Analysis of Randomized Algorithms Tim Roughgarden March 3, 2016 1 Preamble In CS109 and CS161, you learned some tricks of

More information

Threshold Phenomena in Random Constraint Satisfaction Problems. Harold Scott Connamacher

Threshold Phenomena in Random Constraint Satisfaction Problems. Harold Scott Connamacher Threshold Phenomena in Random Constraint Satisfaction Problems by Harold Scott Connamacher A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department

More information

Lecture 4. 1 FPTAS - Fully Polynomial Time Approximation Scheme

Lecture 4. 1 FPTAS - Fully Polynomial Time Approximation Scheme Theory of Computer Science to Msc Students, Spring 2007 Lecturer: Dorit Aharonov Lecture 4 Scribe: Ram Bouobza & Yair Yarom Revised: Shahar Dobzinsi, March 2007 1 FPTAS - Fully Polynomial Time Approximation

More information

Lower Bounds for Testing Bipartiteness in Dense Graphs

Lower Bounds for Testing Bipartiteness in Dense Graphs Lower Bounds for Testing Bipartiteness in Dense Graphs Andrej Bogdanov Luca Trevisan Abstract We consider the problem of testing bipartiteness in the adjacency matrix model. The best known algorithm, due

More information

Lecture 18: Inapproximability of MAX-3-SAT

Lecture 18: Inapproximability of MAX-3-SAT CS 880: Advanced Complexity Theory 3/7/2008 Lecture 18: Inapproximability of MAX-3-SAT Instructor: Dieter van Melkebeek Scribe: Jeff Kinne In this lecture we prove a tight inapproximability result for

More information

NP-Complete Reductions 2

NP-Complete Reductions 2 x 1 x 1 x 2 x 2 x 3 x 3 x 4 x 4 12 22 32 CS 447 11 13 21 23 31 33 Algorithms NP-Complete Reductions 2 Prof. Gregory Provan Department of Computer Science University College Cork 1 Lecture Outline NP-Complete

More information

Topic: Sampling, Medians of Means method and DNF counting Date: October 6, 2004 Scribe: Florin Oprea

Topic: Sampling, Medians of Means method and DNF counting Date: October 6, 2004 Scribe: Florin Oprea 15-859(M): Randomized Algorithms Lecturer: Shuchi Chawla Topic: Sampling, Medians of Means method and DNF counting Date: October 6, 200 Scribe: Florin Oprea 8.1 Introduction In this lecture we will consider

More information

Comp487/587 - Boolean Formulas

Comp487/587 - Boolean Formulas Comp487/587 - Boolean Formulas 1 Logic and SAT 1.1 What is a Boolean Formula Logic is a way through which we can analyze and reason about simple or complicated events. In particular, we are interested

More information

Phase transition for Local Search on planted SAT

Phase transition for Local Search on planted SAT Phase transition for Local Search on planted SAT Andrei A. Bulatov 1 Evgeny S. Skvortsov 2 1 Simon Fraser University 2 Google Canada abulatov@cs.sfu.ca, evgenys@google.com Abstract. The Local Search algorithm

More information

Lecture 15: A Brief Look at PCP

Lecture 15: A Brief Look at PCP IAS/PCMI Summer Session 2000 Clay Mathematics Undergraduate Program Basic Course on Computational Complexity Lecture 15: A Brief Look at PCP David Mix Barrington and Alexis Maciel August 4, 2000 1. Overview

More information

Lecture 2: January 18

Lecture 2: January 18 CS271 Randomness & Computation Spring 2018 Instructor: Alistair Sinclair Lecture 2: January 18 Disclaimer: These notes have not been subjected to the usual scrutiny accorded to formal publications. They

More information

Lecture 5: The Principle of Deferred Decisions. Chernoff Bounds

Lecture 5: The Principle of Deferred Decisions. Chernoff Bounds Randomized Algorithms Lecture 5: The Principle of Deferred Decisions. Chernoff Bounds Sotiris Nikoletseas Associate Professor CEID - ETY Course 2013-2014 Sotiris Nikoletseas, Associate Professor Randomized

More information

Integer Linear Programs

Integer Linear Programs Lecture 2: Review, Linear Programming Relaxations Today we will talk about expressing combinatorial problems as mathematical programs, specifically Integer Linear Programs (ILPs). We then see what happens

More information

UC Berkeley CS 170: Efficient Algorithms and Intractable Problems Handout 22 Lecturer: David Wagner April 24, Notes 22 for CS 170

UC Berkeley CS 170: Efficient Algorithms and Intractable Problems Handout 22 Lecturer: David Wagner April 24, Notes 22 for CS 170 UC Berkeley CS 170: Efficient Algorithms and Intractable Problems Handout 22 Lecturer: David Wagner April 24, 2003 Notes 22 for CS 170 1 NP-completeness of Circuit-SAT We will prove that the circuit satisfiability

More information

Lecture 19: November 10

Lecture 19: November 10 CS294 Markov Chain Monte Carlo: Foundations & Applications Fall 2009 Lecture 19: November 10 Lecturer: Prof. Alistair Sinclair Scribes: Kevin Dick and Tanya Gordeeva Disclaimer: These notes have not been

More information

Lecture 6,7 (Sept 27 and 29, 2011 ): Bin Packing, MAX-SAT

Lecture 6,7 (Sept 27 and 29, 2011 ): Bin Packing, MAX-SAT ,7 CMPUT 675: Approximation Algorithms Fall 2011 Lecture 6,7 (Sept 27 and 29, 2011 ): Bin Pacing, MAX-SAT Lecturer: Mohammad R. Salavatipour Scribe: Weitian Tong 6.1 Bin Pacing Problem Recall the bin pacing

More information

Lecture 10: Hardness of approximating clique, FGLSS graph

Lecture 10: Hardness of approximating clique, FGLSS graph CSE 533: The PCP Theorem and Hardness of Approximation (Autumn 2005) Lecture 10: Hardness of approximating clique, FGLSS graph Nov. 2, 2005 Lecturer: Venkat Guruswami and Ryan O Donnell Scribe: Ioannis

More information

Discrete Mathematics and Probability Theory Fall 2011 Rao Midterm 2 Solutions

Discrete Mathematics and Probability Theory Fall 2011 Rao Midterm 2 Solutions CS 70 Discrete Mathematics and Probability Theory Fall 20 Rao Midterm 2 Solutions True/False. [24 pts] Circle one of the provided answers please! No negative points will be assigned for incorrect answers.

More information

NP-Completeness Review

NP-Completeness Review CS124 NP-Completeness Review Where We Are Headed Up to this point, we have generally assumed that if we were given a problem, we could find a way to solve it. Unfortunately, as most of you know, there

More information

Lecture 14: October 22

Lecture 14: October 22 CS294 Markov Chain Monte Carlo: Foundations & Applications Fall 2009 Lecture 14: October 22 Lecturer: Alistair Sinclair Scribes: Alistair Sinclair Disclaimer: These notes have not been subjected to the

More information

On the Complexity of the Minimum Independent Set Partition Problem

On the Complexity of the Minimum Independent Set Partition Problem On the Complexity of the Minimum Independent Set Partition Problem T-H. Hubert Chan 1, Charalampos Papamanthou 2, and Zhichao Zhao 1 1 Department of Computer Science the University of Hong Kong {hubert,zczhao}@cs.hku.hk

More information

Lecture 12 Scribe Notes

Lecture 12 Scribe Notes 6.890: Algorithmic Lower Bounds: Fun With Hardness Proofs Fall 04 Prof. Erik Demaine Lecture Scribe Notes Recap For the past two classes, we ve been covering inapproximability, some of its reduction styles,

More information

Asymptotically optimal induced universal graphs

Asymptotically optimal induced universal graphs Asymptotically optimal induced universal graphs Noga Alon Abstract We prove that the minimum number of vertices of a graph that contains every graph on vertices as an induced subgraph is (1+o(1))2 ( 1)/2.

More information

A Duality between Clause Width and Clause Density for SAT

A Duality between Clause Width and Clause Density for SAT A Duality between Clause Width and Clause Density for SAT Chris Calabro, Russell Impagliazzo, Ramamohan Paturi University of California, San Diego 9500 Gilman Drive, La Jolla, CA 92093 ccalabro,russell,paturi

More information

NP Complete Problems. COMP 215 Lecture 20

NP Complete Problems. COMP 215 Lecture 20 NP Complete Problems COMP 215 Lecture 20 Complexity Theory Complexity theory is a research area unto itself. The central project is classifying problems as either tractable or intractable. Tractable Worst

More information

Phase Transitions of Bounded Satisfiability Problems*

Phase Transitions of Bounded Satisfiability Problems* Phase Transitions of Bounded Satisfiability Problems* Delbert D. Bailey, Phokion G. Kolaitis Computer Science Department University of California, Santa Cruz Santa Cruz, CA 95064, U.S.A {dbailey,kolaitis}@cse.ucsc.edu

More information

SAT, NP, NP-Completeness

SAT, NP, NP-Completeness CS 473: Algorithms, Spring 2018 SAT, NP, NP-Completeness Lecture 22 April 13, 2018 Most slides are courtesy Prof. Chekuri Ruta (UIUC) CS473 1 Spring 2018 1 / 57 Part I Reductions Continued Ruta (UIUC)

More information

Chapter 2. Reductions and NP. 2.1 Reductions Continued The Satisfiability Problem (SAT) SAT 3SAT. CS 573: Algorithms, Fall 2013 August 29, 2013

Chapter 2. Reductions and NP. 2.1 Reductions Continued The Satisfiability Problem (SAT) SAT 3SAT. CS 573: Algorithms, Fall 2013 August 29, 2013 Chapter 2 Reductions and NP CS 573: Algorithms, Fall 2013 August 29, 2013 2.1 Reductions Continued 2.1.1 The Satisfiability Problem SAT 2.1.1.1 Propositional Formulas Definition 2.1.1. Consider a set of

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 8 Luca Trevisan September 19, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 8 Luca Trevisan September 19, 2017 U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 8 Luca Trevisan September 19, 2017 Scribed by Luowen Qian Lecture 8 In which we use spectral techniques to find certificates of unsatisfiability

More information

The Probabilistic Method

The Probabilistic Method The Probabilistic Method In Graph Theory Ehssan Khanmohammadi Department of Mathematics The Pennsylvania State University February 25, 2010 What do we mean by the probabilistic method? Why use this method?

More information

Almost all graphs with 2.522n edges are not 3-colorable

Almost all graphs with 2.522n edges are not 3-colorable Almost all graphs with 2.522n edges are not 3-colorable Dimitris Achlioptas optas@cs.toronto.edu Michael Molloy molloy@cs.toronto.edu Department of Computer Science University of Toronto Toronto, Ontario

More information

Lecture 18: More NP-Complete Problems

Lecture 18: More NP-Complete Problems 6.045 Lecture 18: More NP-Complete Problems 1 The Clique Problem a d f c b e g Given a graph G and positive k, does G contain a complete subgraph on k nodes? CLIQUE = { (G,k) G is an undirected graph with

More information

1 Agenda. 2 History. 3 Probabilistically Checkable Proofs (PCPs). Lecture Notes Definitions. PCPs. Approximation Algorithms.

1 Agenda. 2 History. 3 Probabilistically Checkable Proofs (PCPs). Lecture Notes Definitions. PCPs. Approximation Algorithms. CS 221: Computational Complexity Prof. Salil Vadhan Lecture Notes 20 April 12, 2010 Scribe: Jonathan Pines 1 Agenda. PCPs. Approximation Algorithms. PCPs = Inapproximability. 2 History. First, some history

More information

A necessary and sufficient condition for the existence of a spanning tree with specified vertices having large degrees

A necessary and sufficient condition for the existence of a spanning tree with specified vertices having large degrees A necessary and sufficient condition for the existence of a spanning tree with specified vertices having large degrees Yoshimi Egawa Department of Mathematical Information Science, Tokyo University of

More information

20.1 2SAT. CS125 Lecture 20 Fall 2016

20.1 2SAT. CS125 Lecture 20 Fall 2016 CS125 Lecture 20 Fall 2016 20.1 2SAT We show yet another possible way to solve the 2SAT problem. Recall that the input to 2SAT is a logical expression that is the conunction (AND) of a set of clauses,

More information

NP-Complete Problems. More reductions

NP-Complete Problems. More reductions NP-Complete Problems More reductions Definitions P: problems that can be solved in polynomial time (typically in n, size of input) on a deterministic Turing machine Any normal computer simulates a DTM

More information

11.1 Set Cover ILP formulation of set cover Deterministic rounding

11.1 Set Cover ILP formulation of set cover Deterministic rounding CS787: Advanced Algorithms Lecture 11: Randomized Rounding, Concentration Bounds In this lecture we will see some more examples of approximation algorithms based on LP relaxations. This time we will use

More information

A = A U. U [n] P(A U ). n 1. 2 k(n k). k. k=1

A = A U. U [n] P(A U ). n 1. 2 k(n k). k. k=1 Lecture I jacques@ucsd.edu Notation: Throughout, P denotes probability and E denotes expectation. Denote (X) (r) = X(X 1)... (X r + 1) and let G n,p denote the Erdős-Rényi model of random graphs. 10 Random

More information

CS151 Complexity Theory. Lecture 15 May 22, 2017

CS151 Complexity Theory. Lecture 15 May 22, 2017 CS151 Complexity Theory Lecture 15 New topic(s) Optimization problems, Approximation Algorithms, and Probabilistically Checkable Proofs Optimization Problems many hard problems (especially NP-hard) are

More information

A New Lower Bound on the Maximum Number of Satisfied Clauses in Max-SAT and its Algorithmic Applications

A New Lower Bound on the Maximum Number of Satisfied Clauses in Max-SAT and its Algorithmic Applications A New Lower Bound on the Maximum Number of Satisfied Clauses in Max-SAT and its Algorithmic Applications Robert Crowston, Gregory Gutin, Mark Jones, Anders Yeo Department of Computer Science Royal Holloway,

More information

LECTURE 10: REVIEW OF POWER SERIES. 1. Motivation

LECTURE 10: REVIEW OF POWER SERIES. 1. Motivation LECTURE 10: REVIEW OF POWER SERIES By definition, a power series centered at x 0 is a series of the form where a 0, a 1,... and x 0 are constants. For convenience, we shall mostly be concerned with the

More information

U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 2 Luca Trevisan August 29, 2017

U.C. Berkeley CS294: Beyond Worst-Case Analysis Handout 2 Luca Trevisan August 29, 2017 U.C. Berkeley CS94: Beyond Worst-Case Analysis Handout Luca Trevisan August 9, 07 Scribe: Mahshid Montazer Lecture In this lecture, we study the Max Cut problem in random graphs. We compute the probable

More information

Connections between exponential time and polynomial time problem Lecture Notes for CS294

Connections between exponential time and polynomial time problem Lecture Notes for CS294 Connections between exponential time and polynomial time problem Lecture Notes for CS294 Lecturer: Russell Impagliazzo Scribe: Stefan Schneider November 10, 2015 We connect problems that have exponential

More information

Out-colourings of Digraphs

Out-colourings of Digraphs Out-colourings of Digraphs N. Alon J. Bang-Jensen S. Bessy July 13, 2017 Abstract We study vertex colourings of digraphs so that no out-neighbourhood is monochromatic and call such a colouring an out-colouring.

More information

The Interface between P and NP: COL, XOR, NAE, 1-in-, and Horn SAT

The Interface between P and NP: COL, XOR, NAE, 1-in-, and Horn SAT The Interface between P and NP: COL, XOR, NAE, -in-, and Horn SAT Toby Walsh APES Report: APES-37-2002 Abstract We study in detail the interface between P and NP by means of five new problem classes. Like

More information

Lecture Notes Each circuit agrees with M on inputs of length equal to its index, i.e. n, x {0, 1} n, C n (x) = M(x).

Lecture Notes Each circuit agrees with M on inputs of length equal to its index, i.e. n, x {0, 1} n, C n (x) = M(x). CS 221: Computational Complexity Prof. Salil Vadhan Lecture Notes 4 February 3, 2010 Scribe: Jonathan Pines 1 Agenda P-/NP- Completeness NP-intermediate problems NP vs. co-np L, NL 2 Recap Last time, we

More information

The chromatic number of random regular graphs

The chromatic number of random regular graphs The chromatic number of random regular graphs Xavier Pérez Giménez Stochastics Numerics Seminar Models of random graphs Erdős & Rényi classic models (G n,p, G n,m ) Fixed degree sequence Regular graphs

More information

Lecture 24: Approximate Counting

Lecture 24: Approximate Counting CS 710: Complexity Theory 12/1/2011 Lecture 24: Approximate Counting Instructor: Dieter van Melkebeek Scribe: David Guild and Gautam Prakriya Last time we introduced counting problems and defined the class

More information

CSE 312 Final Review: Section AA

CSE 312 Final Review: Section AA CSE 312 TAs December 8, 2011 General Information General Information Comprehensive Midterm General Information Comprehensive Midterm Heavily weighted toward material after the midterm Pre-Midterm Material

More information

Not all counting problems are efficiently approximable. We open with a simple example.

Not all counting problems are efficiently approximable. We open with a simple example. Chapter 7 Inapproximability Not all counting problems are efficiently approximable. We open with a simple example. Fact 7.1. Unless RP = NP there can be no FPRAS for the number of Hamilton cycles in a

More information

2 P vs. NP and Diagonalization

2 P vs. NP and Diagonalization 2 P vs NP and Diagonalization CS 6810 Theory of Computing, Fall 2012 Instructor: David Steurer (sc2392) Date: 08/28/2012 In this lecture, we cover the following topics: 1 3SAT is NP hard; 2 Time hierarchies;

More information

Random Bernstein-Markov factors

Random Bernstein-Markov factors Random Bernstein-Markov factors Igor Pritsker and Koushik Ramachandran October 20, 208 Abstract For a polynomial P n of degree n, Bernstein s inequality states that P n n P n for all L p norms on the unit

More information

6.842 Randomness and Computation Lecture 5

6.842 Randomness and Computation Lecture 5 6.842 Randomness and Computation 2012-02-22 Lecture 5 Lecturer: Ronitt Rubinfeld Scribe: Michael Forbes 1 Overview Today we will define the notion of a pairwise independent hash function, and discuss its

More information

Keywords. Approximation Complexity Theory: historical remarks. What s PCP?

Keywords. Approximation Complexity Theory: historical remarks. What s PCP? Keywords The following terms should be well-known to you. 1. P, NP, NP-hard, NP-complete. 2. (polynomial-time) reduction. 3. Cook-Levin theorem. 4. NPO problems. Instances. Solutions. For a long time it

More information

CSC 5170: Theory of Computational Complexity Lecture 9 The Chinese University of Hong Kong 15 March 2010

CSC 5170: Theory of Computational Complexity Lecture 9 The Chinese University of Hong Kong 15 March 2010 CSC 5170: Theory of Computational Complexity Lecture 9 The Chinese University of Hong Kong 15 March 2010 We now embark on a study of computational classes that are more general than NP. As these classes

More information

Lecture 3: Oct 7, 2014

Lecture 3: Oct 7, 2014 Information and Coding Theory Autumn 04 Lecturer: Shi Li Lecture : Oct 7, 04 Scribe: Mrinalkanti Ghosh In last lecture we have seen an use of entropy to give a tight upper bound in number of triangles

More information

Matchings in hypergraphs of large minimum degree

Matchings in hypergraphs of large minimum degree Matchings in hypergraphs of large minimum degree Daniela Kühn Deryk Osthus Abstract It is well known that every bipartite graph with vertex classes of size n whose minimum degree is at least n/2 contains

More information

NP and NP Completeness

NP and NP Completeness CS 374: Algorithms & Models of Computation, Spring 2017 NP and NP Completeness Lecture 23 April 20, 2017 Chandra Chekuri (UIUC) CS374 1 Spring 2017 1 / 44 Part I NP Chandra Chekuri (UIUC) CS374 2 Spring

More information

Lecture 2: September 8

Lecture 2: September 8 CS294 Markov Chain Monte Carlo: Foundations & Applications Fall 2009 Lecture 2: September 8 Lecturer: Prof. Alistair Sinclair Scribes: Anand Bhaskar and Anindya De Disclaimer: These notes have not been

More information

Nondeterminism LECTURE Nondeterminism as a proof system. University of California, Los Angeles CS 289A Communication Complexity

Nondeterminism LECTURE Nondeterminism as a proof system. University of California, Los Angeles CS 289A Communication Complexity University of California, Los Angeles CS 289A Communication Complexity Instructor: Alexander Sherstov Scribe: Matt Brown Date: January 25, 2012 LECTURE 5 Nondeterminism In this lecture, we introduce nondeterministic

More information

Lecture 2 (Notes) 1. The book Computational Complexity: A Modern Approach by Sanjeev Arora and Boaz Barak;

Lecture 2 (Notes) 1. The book Computational Complexity: A Modern Approach by Sanjeev Arora and Boaz Barak; Topics in Theoretical Computer Science February 29, 2016 Lecturer: Ola Svensson Lecture 2 (Notes) Scribes: Ola Svensson Disclaimer: These notes were written for the lecturer only and may contain inconsistent

More information

Lecture 2 Sep 5, 2017

Lecture 2 Sep 5, 2017 CS 388R: Randomized Algorithms Fall 2017 Lecture 2 Sep 5, 2017 Prof. Eric Price Scribe: V. Orestis Papadigenopoulos and Patrick Rall NOTE: THESE NOTES HAVE NOT BEEN EDITED OR CHECKED FOR CORRECTNESS 1

More information