CS Homework 3. Walid Krichene ( )

Size: px
Start display at page:

Download "CS Homework 3. Walid Krichene ( )"

Transcription

1 CS 70 - Homework 3 Walid Krichene (36517) (1) Consider the randomized multiplicative weights algorithm on the alternatingly correct pair of experts A, B, with update (1 ). Without loss of generality, assume that A is correct on even days, and B on odd days. At iteration t, the loss vector l(t) = [l A (t), l B (t)] is given by { [1, 0] if t is odd l(t) = [0, 1] if t is even therefore the wight updates are given by w(t + 1) = { [w A (t)(1 ), w B (t)] if t is odd [w A (t), w B (t)(1 )] if t is even by induction, stating from the uniform weights w(1) = [1, 1], we have w(k + 1) = [(1 ) k, (1 ) k ] w(k) = [(1 ) k, (1 ) k 1 ] Let L(t) be the (random) loss incurred by the algorithm at iteration t. hen the expected loss at iteration t is E[L(t)] = w A(t)l A (t) + w B (t)l B (t) (1 ) k = 1 (1 ) = k if t = k + 1 w A (t) + w B (t) (1 ) k 1 if t = k Finally, the expected loss over iterations is = 1 (1 ) k +(1 ) k 1 E[L(t)] = / 1 + / 1 since 1/ 1/( ), an upper bound on the expected loss is simply given by E[L(t)] For the case = 1, the expectation becomes E[L(t)] = / 1 + / 3 1

2 () he original matrix of the game is A = In order to have payoffs in [0, 1], an equivalent game is given by the scaled and translated matrix Ã, given by Ãi,j = Ai,j+1 Ã = 1/ / / 1/ 0 0 he experts algorithm applied to the -player zero sum game proceeds as follows: the row player applies the expert s algorithm, and maintains the probability distribution p(t). the outcomes at iteration t are given by l(t) = A.y (t) where y(t) is the best response to p(t) (the column player is assumed to play optimally), given by y(t) = arg max p(t)ay y the column player simply picks arg max j (p(t)a) j, and picks the lowest such j in case of ties. he results are x = 1 x(t) δ/ 10 (0.491, , , 0.491) ( , , , ) ln 4 ( , , , ) where δ/ is computed using the average distributions, and is the maximum increase in player utility if she unilaterally changes her strategy. In other words where δ/ = max( xãȳ R(ȳ), C( x) xãȳ ) x = 1 ȳ = 1 x(t) y(t) and R(y) = min xãy x C(x) = max y xãy

3 1 package e x p e r t s import u t i l. l i n A l g e b r a. 3 4 c l a s s InvalidArgumentException ( message : S t r i n g ) extends Exception ( message ) 5 6 t r a i t Nature { 7 d e f update ( s t r a t e g y : Array [ Double ] ) 8 } 9 10 a b s t r a c t c l a s s Expert [N<: Nature ] ( v a l nature : N) { 11 d e f nextloss ( ) : Double 1 } c l a s s ZeroSumGame ( v a l A: Array [ Array [ Double ] ] ) extends Nature { 15 // A i s the p a y o f f matrix, assumed to be normalized ( e n t r i e s i n [ 0, 1 ] ) 16 // row minimizes, column maximizes 17 v a l nrows = A. l e n g t h 18 v a l ncols = A( 0 ). l e n g t h 19 var rounds = 0 0 var bestcolresponse = 0 1 v a l c u m u l a t i v e C o l S t r a t e g y = new Array [ Double ] ( ncols ) 3 v a l cumulativerowstrategy = new Array [ Double ] ( nrows ) 4 5 p r i v a t e d e f getnormalized ( c u m u l a t i v e S t r a t e g y : Array [ Double ] ) : Array [ Double ] = { 6 v a l norm = c u m u l a t i v e S t r a t e g y. sum 7 c u m u l a t i v e S t r a t e g y. map( /norm ) 8 } 9 30 d e f getavgcolstrategy ( ) = getnormalized ( c u m u l a t i v e C o l S t r a t e g y ) 31 d e f getavgrowstrategy ( ) = getnormalized ( cumulativerowstrategy ) 3 33 d e f computeoutcome ( x : Array [ Double ], y : Array [ Double ] ) : Double = { 34 x. times (A). times ( y. transp ( ) ). doublevalue 35 } d e f computebestcolresponse ( rowstrategy : Array [ Double ] ) : ( Int, Double ) = { 38 // compute the b e s t column r e s p o n s e 39 v a l ( (, argmax ), max) = rowstrategy. times (A). argmax ( ) 40 ( argmax, max) 41 } 4 43 d e f computebestrowresponse ( c o l S t r a t e g y : Array [ Double ] ) : ( Int, Double ) = { 44 v a l ( (, argmin ), min ) = A. times ( c o l S t r a t e g y. transp ( ) ). argmin ( ) 45 ( argmin, min ) 46 } d e f update ( rowstrategy : Array [ Double ] ) { 49 bestcolresponse = computebestcolresponse ( rowstrategy ) rounds += 1 51 // update the average s t r a t e g i e s 5 f o r ( i < 0 to nrows 1) 53 cumulativerowstrategy ( i ) += rowstrategy ( i ) c u m u l a t i v e C o l S t r a t e g y ( bestcolresponse ) += 1 56 } 57 } c l a s s ZeroSumGameExpert ( game : ZeroSumGame, row : I n t ) extends Expert [ ZeroSumGame ] ( game ){ 60 d e f nextloss ( ) : Double = { 61 v a l colresponse = game. bestcolresponse 6 r e t u r n game.a( row ) ( colresponse ) 63 } 64 } 3

4 65 66 // Experts a l g o r i t h m 67 // ype parameter N must extend Nature, and e x p e r t s are o f type Expert [N] 68 // ( the e x p e r t s s h a r e an i n s t a n c e o f Nature ) 69 // nature keeps e x t r a i n f o r m a t i o n s p e c i f i c to the p a r t i c u l a r game being played. 70 // f o r example, i f nature i s a ZeroSumGame, i t p r o v i d e s methods to compute b e s t 71 // column response, and i t keeps t r a c k o f the average row and column s t r a t e g i e s. 7 // In the r o u t i n g game, i t would p r o v i d e a b e s t r e s p o n s e method that computes the 73 // s h o r t e s t path, e t c. 74 c l a s s ExpertAlgorithm [N<: Nature ] ( e p s i l o n : Double, e x p e r t s : L i s t [ Expert [N ] ] ) { 75 i f ( e p s i l o n >= 1 e p s i l o n <= 0) 76 throw new InvalidArgumentException ( e p s i l o n should be i n ( 0, 1) ) v a l nature = e x p e r t s ( 0 ). nature 79 v a l support = e x p e r t s. l e n g t h 80 var s t r a t e g y = Array. f i l l ( support ) ( 1. / support ) 81 8 d e f n o r m a l i z e ( ) { 83 v a l norm = s t r a t e g y. sum 84 f o r ( i < 0 to support 1) 85 s t r a t e g y ( i )/=norm 86 } d e f next ( ) { 89 nature. update ( s t r a t e g y ) 90 v a l weights = e x p e r t s. map(. nextloss ). map( math. pow(1 e p s i l o n, ) ). toarray 91 s t r a t e g y = s t r a t e g y. dotmult ( weights ). arrayvalue ( ) 9 normalize ( ) 93 } 94 } 1 package e x p e r t s 3 o b j e c t main { 4 d e f main ( a r g s : Array [ S t r i n g ] ) : Unit = { 5 v a l e p s i l o n =. 1 6 // p a y o f f matrix 7 v a l A = Array [ Array [ Double ] ] ( 8 Array [ Double ] (. 5, 1, 0 ), 9 Array [ Double ] ( 0,. 5, 1 ), 10 Array [ Double ] ( 1, 0,. 5 ), 11 Array [ Double ] (. 5, 0, 0) 1 ) 13 v a l game = new ZeroSumGame (A) 14 v a l e x p e r t s = (0 to 3 ). map( new ZeroSumGameExpert ( game, ) ). t o L i s t 15 v a l a l g = new ExpertAlgorithm [ ZeroSumGame ] ( e p s i l o n, e x p e r t s ) f o r ( t < 1 to math. c e i l (100 math. l o g ( 4 ) ). intvalue ) 18 a l g. next ( ) 19 0 v a l y = game. getavgcolstrategy ( ) 1 v a l x = game. getavgrowstrategy ( ) v a l d e l t a = math. max( 3 game. computebestcolresponse ( x ). game. computeoutcome ( x, y ), 4 game. computeoutcome ( x, y ) game. computebestrowresponse ( y ). ) 5 6 p r i n t l n ( game. rounds ) 7 p r i n t l n ( x. t o L i s t ) 8 p r i n t l n ( d e l t a ) 9 } 30 } he code is in the files experts.scala and main.scala, and some helper classes are in linalgebra.scala. 4

5 (3) Consider the fractional routing problem on the graph G = (V, E), with pairs (s i, t i ), i [k]. Let E = m and V = n. Find a 1 + approximate solution to the fractional path routing problem in O(km log m ) time (with high probability). Lemma 1 If one chooses uniformly at random from k possibilities, times, with i [k], if i is the number of times i is chosen, we have P ( i k ) (1 ) > 1 1 k 8k ln k, then for each Lemma If G = G t, where the choice of G t is made independently of others, and G t have mean and covariance at most k, then 8k ln k, and P(G E[G](1 + )) > 1 1 k answer Consider the following randomized version of the experts algorithm, where the toll player maintains a probability distribution p e (t), e E. Algorithm 1 Experts algorithm Initialize i := 0 i for each iteration t {1,..., } do the routing player chooses, uniformly at random, i [k], and routes (s i, t i ) along the shortest path in the metric given by p e (t) let f i (t) F i be the corresponding flow solution (here F i is the set of feasible flows that connect (s i, t i )). let f j (t) = 0 for all j i increment i the toll player updates the distribution using the experts update rule p e (t + 1) p e (t)(1 + ) ge(t) where the gain of expert e on day t is the congestion of edge e under flow f(t) g e (t) = c(e, f(t)) Note: since the routing player only routes one pair (s i, t i ) on each day, the gains are at most one: g e (t) 1 for all e and t end for the candidate approximate flow solution is k 1 f = i=1 i f i (t) ( f is a random variable) Let 5

6 c max ( f) = max e E c(e, f) be the maximal congestion corresponding to f (random variable) c be the value of the toll congestion game c = min f F [k] max c(p, f) = max min c(p, f) p E f F [k] where p is in the set of probability distributions over the set of edges E, and f is in the set of feasible flows that connect all k pairs (F [k] = F F k = {f f k, f i F i ). Note: for each fixed p, the congestion function f c(p, f) is linear. his is crucial for the proof. G be the expected total gain G = p E c(p(t), f(t)) (random variable since the f(t) s are random). c(p(t), f(t)) is the cost of the shortest path picked on day t. G be the gain of the best expert (random variable) G = max c(e, f(t)) e 8k log(m) and assume we play the game for = days. he complexity of each iteration is O(m log(m)) and corresponds to computing one shortest path for the randomly chosen pair. hus the total time complexity is O( m log(m)) = O(km log(m) ). he claim is that with high probability, f is a (1 + 3)-approximate solution, i.e. 1 cmax ( f) c (1 + 3) (here c max ( f) is greater than the value of the game since the toll player plays second) or equivalently Proof of claim: we use a sequence of inequalities: c (1 3) c max ( f) 1. by the experts algorithm analysis, we have that for every realization, G (1 )G log(m)/, thus using = 8k log m/, we have kg. with high probability (probability 1 1 k ), t (1 )kg k log(m) = (1 ) kg 8 kg (1 )cmax ( f) () Indeed: by definition, G = max e t c(e, f(t)). Fixing e, we can write c(e, f(t)) = c(e, f i (t)) t t i = 1 i c(e, f i (t)) i i t (1 ) 1 c(e, f i (t)) using i (1 ) (Lemma 1) k i k i t = (1 ) k c(e, f) by linearity of c(e,.) taking the maximum over e, we have k max e t c(e, f(t)) (1 ) max e c(e, f), k i.e. G (1 )c max ( f) 6 (1)

7 3. with high probability (probability 1 1 k ), c (1 ) k G (3) Indeed, we have G = c(p(t), f(t)), where f(t) = f i(t) for some i, and the choice of i is made independently and uniformly at random. We have E[c(p(t), f(t))] = 1 c(p(t), k E[f(t)]) k by linearity. Here we observe that E[f(t)] = k i=1 1 k f i(t) (each i is chosen with probability 1/k), therefore k E[f(t)] = k i=1 f i(t) is the solution to the k-shortest paths, routing all pairs (since each f i (t) is the shortest path routing pair i). In other words, c(p(t), k E[f(t)]) = min f F[k] c(p(t), f), and thus is less than c (c = max p min f ). Now c c(p(t), E[f(t)]) = k E[c(p(t), f(t))] (where each c(p(t), f(t)) has expectation and variance less than one). Summing over t, c k E[G] then using Lemma, we have with high probability, E[G] G(1 ), therefore c k E[G] G(1 ) Combining these inequalities, we have with high probability c (1 ) kg by (3) (1 )((1 ) kg ) by (1) 8 (1 ) kg 8 (1 ) 3 c max ( f) 8 by () (1 3)c max ( f) 8 (there is a /8 additive term here, but I assume it is okay since we had a similar term in the original algorithm) 7

8 (4) Consider investing the following scenario. here are n possible instruments one can invest in in a period of time intervals, where each instrument either pays one of 0 dollars or 1 dollar at each time. You have inside information that at least one of the instruments yields P dollars over the time intervals. One can only invest in one instrument. he problem is to decide which one to invest in and when to do so. (a) Show that any deterministic algorithm may yield zero dollars (unless n = 1) proof Let us denote the gain of instrument i on time t by g i (t). Consider a deterministic algorithm A, and consider the sequence g perfect of gains such that g perfect i (t) = 1 i and t. Let i 0, t 0 be, respectively, the instrument invested in and the time invested in, when A is run on g perfect. Since A is deterministic (and causal!) when we run A on any sequence g such that g i (t) = 1 i and t < t 0, the algorithm will invest in i 0 on t 0. In particular if run on g haha defined by { gi haha 0 if i = i 0 and t t 0 (t) = 1 otherwise the gain is 0. (note that the sequence is feasible since all instruments except i 0 have maximal gain) 8

COS 511: Theoretical Machine Learning. Lecturer: Rob Schapire Lecture 24 Scribe: Sachin Ravi May 2, 2013

COS 511: Theoretical Machine Learning. Lecturer: Rob Schapire Lecture 24 Scribe: Sachin Ravi May 2, 2013 COS 5: heoretical Machine Learning Lecturer: Rob Schapire Lecture 24 Scribe: Sachin Ravi May 2, 203 Review of Zero-Sum Games At the end of last lecture, we discussed a model for two player games (call

More information

Learning, Games, and Networks

Learning, Games, and Networks Learning, Games, and Networks Abhishek Sinha Laboratory for Information and Decision Systems MIT ML Talk Series @CNRG December 12, 2016 1 / 44 Outline 1 Prediction With Experts Advice 2 Application to

More information

princeton univ. F 13 cos 521: Advanced Algorithm Design Lecture 17: Duality and MinMax Theorem Lecturer: Sanjeev Arora

princeton univ. F 13 cos 521: Advanced Algorithm Design Lecture 17: Duality and MinMax Theorem Lecturer: Sanjeev Arora princeton univ F 13 cos 521: Advanced Algorithm Design Lecture 17: Duality and MinMax Theorem Lecturer: Sanjeev Arora Scribe: Today we first see LP duality, which will then be explored a bit more in the

More information

COS 511: Theoretical Machine Learning. Lecturer: Rob Schapire Lecture #24 Scribe: Jad Bechara May 2, 2018

COS 511: Theoretical Machine Learning. Lecturer: Rob Schapire Lecture #24 Scribe: Jad Bechara May 2, 2018 COS 5: heoretical Machine Learning Lecturer: Rob Schapire Lecture #24 Scribe: Jad Bechara May 2, 208 Review of Game heory he games we will discuss are two-player games that can be modeled by a game matrix

More information

Online Learning, Mistake Bounds, Perceptron Algorithm

Online Learning, Mistake Bounds, Perceptron Algorithm Online Learning, Mistake Bounds, Perceptron Algorithm 1 Online Learning So far the focus of the course has been on batch learning, where algorithms are presented with a sample of training data, from which

More information

Finite Math Section 4_1 Solutions and Hints

Finite Math Section 4_1 Solutions and Hints Finite Math Section 4_1 Solutions and Hints by Brent M. Dingle for the book: Finite Mathematics, 7 th Edition by S. T. Tan. DO NOT PRINT THIS OUT AND TURN IT IN!!!!!!!! This is designed to assist you in

More information

1 Overview. 2 Learning from Experts. 2.1 Defining a meaningful benchmark. AM 221: Advanced Optimization Spring 2016

1 Overview. 2 Learning from Experts. 2.1 Defining a meaningful benchmark. AM 221: Advanced Optimization Spring 2016 AM 1: Advanced Optimization Spring 016 Prof. Yaron Singer Lecture 11 March 3rd 1 Overview In this lecture we will introduce the notion of online convex optimization. This is an extremely useful framework

More information

6. Linear Programming

6. Linear Programming Linear Programming 6-1 6. Linear Programming Linear Programming LP reduction Duality Max-flow min-cut, Zero-sum game Integer Programming and LP relaxation Maximum Bipartite Matching, Minimum weight vertex

More information

M340(921) Solutions Practice Problems (c) 2013, Philip D Loewen

M340(921) Solutions Practice Problems (c) 2013, Philip D Loewen M340(921) Solutions Practice Problems (c) 2013, Philip D Loewen 1. Consider a zero-sum game between Claude and Rachel in which the following matrix shows Claude s winnings: C 1 C 2 C 3 R 1 4 2 5 R G =

More information

CS261: A Second Course in Algorithms Lecture #12: Applications of Multiplicative Weights to Games and Linear Programs

CS261: A Second Course in Algorithms Lecture #12: Applications of Multiplicative Weights to Games and Linear Programs CS26: A Second Course in Algorithms Lecture #2: Applications of Multiplicative Weights to Games and Linear Programs Tim Roughgarden February, 206 Extensions of the Multiplicative Weights Guarantee Last

More information

CS261: A Second Course in Algorithms Lecture #11: Online Learning and the Multiplicative Weights Algorithm

CS261: A Second Course in Algorithms Lecture #11: Online Learning and the Multiplicative Weights Algorithm CS61: A Second Course in Algorithms Lecture #11: Online Learning and the Multiplicative Weights Algorithm Tim Roughgarden February 9, 016 1 Online Algorithms This lecture begins the third module of the

More information

Lecture 16: Perceptron and Exponential Weights Algorithm

Lecture 16: Perceptron and Exponential Weights Algorithm EECS 598-005: Theoretical Foundations of Machine Learning Fall 2015 Lecture 16: Perceptron and Exponential Weights Algorithm Lecturer: Jacob Abernethy Scribes: Yue Wang, Editors: Weiqing Yu and Andrew

More information

REVISION SHEET DECISION MATHS 2 DECISION ANALYSIS

REVISION SHEET DECISION MATHS 2 DECISION ANALYSIS REVISION SHEET DECISION MATHS 2 DECISION ANALYSIS The main ideas are covered in AQA Edexcel MEI OCR D2 Before the exam you should know The meaning of the different kinds of node. Be able to construct a

More information

0.1 Motivating example: weighted majority algorithm

0.1 Motivating example: weighted majority algorithm princeton univ. F 16 cos 521: Advanced Algorithm Design Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm Lecturer: Sanjeev Arora Scribe: Sanjeev Arora (Today s notes

More information

Today: Linear Programming (con t.)

Today: Linear Programming (con t.) Today: Linear Programming (con t.) COSC 581, Algorithms April 10, 2014 Many of these slides are adapted from several online sources Reading Assignments Today s class: Chapter 29.4 Reading assignment for

More information

Ph.D. Qualifying Exam Monday Tuesday, January 4 5, 2016

Ph.D. Qualifying Exam Monday Tuesday, January 4 5, 2016 Ph.D. Qualifying Exam Monday Tuesday, January 4 5, 2016 Put your solution to each problem on a separate sheet of paper. Problem 1. (5106) Find the maximum likelihood estimate of θ where θ is a parameter

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

Routing Games 1. Sandip Chakraborty. Department of Computer Science and Engineering, INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR.

Routing Games 1. Sandip Chakraborty. Department of Computer Science and Engineering, INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR. Routing Games 1 Sandip Chakraborty Department of Computer Science and Engineering, INDIAN INSTITUTE OF TECHNOLOGY KHARAGPUR November 5, 2015 1 Source: Routing Games by Tim Roughgarden Sandip Chakraborty

More information

SF2972 Game Theory Exam with Solutions March 15, 2013

SF2972 Game Theory Exam with Solutions March 15, 2013 SF2972 Game Theory Exam with s March 5, 203 Part A Classical Game Theory Jörgen Weibull and Mark Voorneveld. (a) What are N, S and u in the definition of a finite normal-form (or, equivalently, strategic-form)

More information

Game Theory, On-line prediction and Boosting (Freund, Schapire)

Game Theory, On-line prediction and Boosting (Freund, Schapire) Game heory, On-line prediction and Boosting (Freund, Schapire) Idan Attias /4/208 INRODUCION he purpose of this paper is to bring out the close connection between game theory, on-line prediction and boosting,

More information

1 Machine Learning Concepts (16 points)

1 Machine Learning Concepts (16 points) CSCI 567 Fall 2018 Midterm Exam DO NOT OPEN EXAM UNTIL INSTRUCTED TO DO SO PLEASE TURN OFF ALL CELL PHONES Problem 1 2 3 4 5 6 Total Max 16 10 16 42 24 12 120 Points Please read the following instructions

More information

Informed Principal in Private-Value Environments

Informed Principal in Private-Value Environments Informed Principal in Private-Value Environments Tymofiy Mylovanov Thomas Tröger University of Bonn June 21, 2008 1/28 Motivation 2/28 Motivation In most applications of mechanism design, the proposer

More information

Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm. Lecturer: Sanjeev Arora

Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm. Lecturer: Sanjeev Arora princeton univ. F 13 cos 521: Advanced Algorithm Design Lecture 8: Decision-making under total uncertainty: the multiplicative weight algorithm Lecturer: Sanjeev Arora Scribe: (Today s notes below are

More information

Vickrey Auction. Mechanism Design. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Mechanism Design

Vickrey Auction. Mechanism Design. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Mechanism Design Algorithmic Game Theory Vickrey Auction Vickrey-Clarke-Groves Mechanisms Mechanisms with Money Player preferences are quantifiable. Common currency enables utility transfer between players. Preference

More information

Exact and Approximate Equilibria for Optimal Group Network Formation

Exact and Approximate Equilibria for Optimal Group Network Formation Exact and Approximate Equilibria for Optimal Group Network Formation Elliot Anshelevich and Bugra Caskurlu Computer Science Department, RPI, 110 8th Street, Troy, NY 12180 {eanshel,caskub}@cs.rpi.edu Abstract.

More information

15-850: Advanced Algorithms CMU, Fall 2018 HW #4 (out October 17, 2018) Due: October 28, 2018

15-850: Advanced Algorithms CMU, Fall 2018 HW #4 (out October 17, 2018) Due: October 28, 2018 15-850: Advanced Algorithms CMU, Fall 2018 HW #4 (out October 17, 2018) Due: October 28, 2018 Usual rules. :) Exercises 1. Lots of Flows. Suppose you wanted to find an approximate solution to the following

More information

STAT 414: Introduction to Probability Theory

STAT 414: Introduction to Probability Theory STAT 414: Introduction to Probability Theory Spring 2016; Homework Assignments Latest updated on April 29, 2016 HW1 (Due on Jan. 21) Chapter 1 Problems 1, 8, 9, 10, 11, 18, 19, 26, 28, 30 Theoretical Exercises

More information

Game Theory. Wolfgang Frimmel. Perfect Bayesian Equilibrium

Game Theory. Wolfgang Frimmel. Perfect Bayesian Equilibrium Game Theory Wolfgang Frimmel Perfect Bayesian Equilibrium / 22 Bayesian Nash equilibrium and dynamic games L M R 3 2 L R L R 2 2 L R L 2,, M,2, R,3,3 2 NE and 2 SPNE (only subgame!) 2 / 22 Non-credible

More information

Advanced Topics in Machine Learning and Algorithmic Game Theory Fall semester, 2011/12

Advanced Topics in Machine Learning and Algorithmic Game Theory Fall semester, 2011/12 Advanced Topics in Machine Learning and Algorithmic Game Theory Fall semester, 2011/12 Lecture 4: Multiarmed Bandit in the Adversarial Model Lecturer: Yishay Mansour Scribe: Shai Vardi 4.1 Lecture Overview

More information

1 Primals and Duals: Zero Sum Games

1 Primals and Duals: Zero Sum Games CS 124 Section #11 Zero Sum Games; NP Completeness 4/15/17 1 Primals and Duals: Zero Sum Games We can represent various situations of conflict in life in terms of matrix games. For example, the game shown

More information

Equations and Inequalities

Equations and Inequalities Equations and Inequalities 2 Figure 1 CHAPTER OUTLINE 2.1 The Rectangular Coordinate Systems and Graphs 2.2 Linear Equations in One Variable 2.3 Models and Applications 2.4 Complex Numbers 2.5 Quadratic

More information

Statistical Data Analysis

Statistical Data Analysis DS-GA 0 Lecture notes 8 Fall 016 1 Descriptive statistics Statistical Data Analysis In this section we consider the problem of analyzing a set of data. We describe several techniques for visualizing the

More information

Theoretical Computer Science

Theoretical Computer Science Theoretical Computer Science 0 (009) 599 606 Contents lists available at ScienceDirect Theoretical Computer Science journal homepage: www.elsevier.com/locate/tcs Polynomial algorithms for approximating

More information

1 Markov decision processes

1 Markov decision processes 2.997 Decision-Making in Large-Scale Systems February 4 MI, Spring 2004 Handout #1 Lecture Note 1 1 Markov decision processes In this class we will study discrete-time stochastic systems. We can describe

More information

Lecture notes for quantum semidefinite programming (SDP) solvers

Lecture notes for quantum semidefinite programming (SDP) solvers CMSC 657, Intro to Quantum Information Processing Lecture on November 15 and 0, 018 Fall 018, University of Maryland Prepared by Tongyang Li, Xiaodi Wu Lecture notes for quantum semidefinite programming

More information

LEARNING IN CONCAVE GAMES

LEARNING IN CONCAVE GAMES LEARNING IN CONCAVE GAMES P. Mertikopoulos French National Center for Scientific Research (CNRS) Laboratoire d Informatique de Grenoble GSBE ETBC seminar Maastricht, October 22, 2015 Motivation and Preliminaries

More information

CO759: Algorithmic Game Theory Spring 2015

CO759: Algorithmic Game Theory Spring 2015 CO759: Algorithmic Game Theory Spring 2015 Instructor: Chaitanya Swamy Assignment 1 Due: By Jun 25, 2015 You may use anything proved in class directly. I will maintain a FAQ about the assignment on the

More information

STAT 418: Probability and Stochastic Processes

STAT 418: Probability and Stochastic Processes STAT 418: Probability and Stochastic Processes Spring 2016; Homework Assignments Latest updated on April 29, 2016 HW1 (Due on Jan. 21) Chapter 1 Problems 1, 8, 9, 10, 11, 18, 19, 26, 28, 30 Theoretical

More information

Algorithmic Game Theory. Alexander Skopalik

Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory Alexander Skopalik Today Course Mechanics & Overview Introduction into game theory and some examples Chapter 1: Selfish routing Alexander Skopalik Skopalik@mail.uni-paderborn.de

More information

Algorithms and Data Structures 2014 Exercises week 5

Algorithms and Data Structures 2014 Exercises week 5 Algorithms and Data Structures 014 Exercises week 5 October, 014 Exercises marked by ( ) are hard, but they might show up on the exam. Exercises marked by ( ) are even harder, but they will not be on the

More information

PART 1: USING SCIENTIFIC CALCULATORS (41 PTS.) 1) The Vertex Form for the equation of a parabola in the usual xy-plane is given by y = 3 x + 4

PART 1: USING SCIENTIFIC CALCULATORS (41 PTS.) 1) The Vertex Form for the equation of a parabola in the usual xy-plane is given by y = 3 x + 4 MIDTERM SOLUTIONS (CHAPTERS AND 3: POLYNOMIAL, RATIONAL, EXP L, LOG FUNCTIONS) MATH 141 FALL 018 KUNIYUKI 150 POINTS TOTAL: 41 FOR PART 1, AND 109 FOR PART PART 1: USING SCIENTIFIC CALCULATORS (41 PTS.)

More information

Ambiguity and the Centipede Game

Ambiguity and the Centipede Game Ambiguity and the Centipede Game Jürgen Eichberger, Simon Grant and David Kelsey Heidelberg University, Australian National University, University of Exeter. University of Exeter. June 2018 David Kelsey

More information

Game Theory: Spring 2017

Game Theory: Spring 2017 Game Theory: Spring 207 Ulle Endriss Institute for Logic, Language and Computation University of Amsterdam Ulle Endriss Plan for Today We have seen that every normal-form game has a Nash equilibrium, although

More information

CS364A: Algorithmic Game Theory Lecture #13: Potential Games; A Hierarchy of Equilibria

CS364A: Algorithmic Game Theory Lecture #13: Potential Games; A Hierarchy of Equilibria CS364A: Algorithmic Game Theory Lecture #13: Potential Games; A Hierarchy of Equilibria Tim Roughgarden November 4, 2013 Last lecture we proved that every pure Nash equilibrium of an atomic selfish routing

More information

Gambling in a rigged casino: The adversarial multi-armed bandit problem

Gambling in a rigged casino: The adversarial multi-armed bandit problem Gambling in a rigged casino: The adversarial multi-armed bandit problem Peter Auer Institute for Theoretical Computer Science University of Technology Graz A-8010 Graz (Austria) pauer@igi.tu-graz.ac.at

More information

ACO Comprehensive Exam 19 March Graph Theory

ACO Comprehensive Exam 19 March Graph Theory 1. Graph Theory Let G be a connected simple graph that is not a cycle and is not complete. Prove that there exist distinct non-adjacent vertices u, v V (G) such that the graph obtained from G by deleting

More information

From Bandits to Experts: A Tale of Domination and Independence

From Bandits to Experts: A Tale of Domination and Independence From Bandits to Experts: A Tale of Domination and Independence Nicolò Cesa-Bianchi Università degli Studi di Milano N. Cesa-Bianchi (UNIMI) Domination and Independence 1 / 1 From Bandits to Experts: A

More information

CS286.2 Lecture 8: A variant of QPCP for multiplayer entangled games

CS286.2 Lecture 8: A variant of QPCP for multiplayer entangled games CS286.2 Lecture 8: A variant of QPCP for multiplayer entangled games Scribe: Zeyu Guo In the first lecture, we saw three equivalent variants of the classical PCP theorems in terms of CSP, proof checking,

More information

Multiple Equilibria in the Citizen-Candidate Model of Representative Democracy.

Multiple Equilibria in the Citizen-Candidate Model of Representative Democracy. Multiple Equilibria in the Citizen-Candidate Model of Representative Democracy. Amrita Dhillon and Ben Lockwood This version: March 2001 Abstract De Sinopoli and Turrini (1999) present an example to show

More information

South Pacific Form Seven Certificate

South Pacific Form Seven Certificate 141/1 South Pacific Form Seven Certificate INSTRUCTIONS MATHEMATICS WITH STATISTICS 2015 QUESTION and ANSWER BOOKLET Time allowed: Two and a half hours Write your Student Personal Identification Number

More information

Rutgers Business School, Introduction to Probability, 26:960:575. Homework 1. Prepared by Daniel Pirutinsky. Feburary 1st, 2016

Rutgers Business School, Introduction to Probability, 26:960:575. Homework 1. Prepared by Daniel Pirutinsky. Feburary 1st, 2016 Rutgers usiness School, Introduction to Probability, 6:960:575 Homework Prepared by Daniel Pirutinsky Feburary st, 06 Introduction Suppose we have two boxes, labeled and. Each box contains some amount

More information

When to Ask for an Update: Timing in Strategic Communication

When to Ask for an Update: Timing in Strategic Communication When to Ask for an Update: Timing in Strategic Communication Work in Progress Ying Chen Johns Hopkins University Atara Oliver Rice University March 19, 2018 Main idea In many communication situations,

More information

Optimization 4. GAME THEORY

Optimization 4. GAME THEORY Optimization GAME THEORY DPK Easter Term Saddle points of two-person zero-sum games We consider a game with two players Player I can choose one of m strategies, indexed by i =,, m and Player II can choose

More information

Notes from Week 8: Multi-Armed Bandit Problems

Notes from Week 8: Multi-Armed Bandit Problems CS 683 Learning, Games, and Electronic Markets Spring 2007 Notes from Week 8: Multi-Armed Bandit Problems Instructor: Robert Kleinberg 2-6 Mar 2007 The multi-armed bandit problem The multi-armed bandit

More information

CS 598RM: Algorithmic Game Theory, Spring Practice Exam Solutions

CS 598RM: Algorithmic Game Theory, Spring Practice Exam Solutions CS 598RM: Algorithmic Game Theory, Spring 2017 1. Answer the following. Practice Exam Solutions Agents 1 and 2 are bargaining over how to split a dollar. Each agent simultaneously demands share he would

More information

Game Theory and its Applications to Networks - Part I: Strict Competition

Game Theory and its Applications to Networks - Part I: Strict Competition Game Theory and its Applications to Networks - Part I: Strict Competition Corinne Touati Master ENS Lyon, Fall 200 What is Game Theory and what is it for? Definition (Roger Myerson, Game Theory, Analysis

More information

Exact and Approximate Equilibria for Optimal Group Network Formation

Exact and Approximate Equilibria for Optimal Group Network Formation Noname manuscript No. will be inserted by the editor) Exact and Approximate Equilibria for Optimal Group Network Formation Elliot Anshelevich Bugra Caskurlu Received: December 2009 / Accepted: Abstract

More information

Re: January 27, 2015 Math 080: Final Exam Review Page 1 of 6

Re: January 27, 2015 Math 080: Final Exam Review Page 1 of 6 Re: January 7, 015 Math 080: Final Exam Review Page 1 of 6 Note: If you have difficulty with any of these problems, get help, then go back to the appropriate sections and work more problems! 1. Solve for

More information

Lecture notes for Analysis of Algorithms : Markov decision processes

Lecture notes for Analysis of Algorithms : Markov decision processes Lecture notes for Analysis of Algorithms : Markov decision processes Lecturer: Thomas Dueholm Hansen June 6, 013 Abstract We give an introduction to infinite-horizon Markov decision processes (MDPs) with

More information

CS 781 Lecture 9 March 10, 2011 Topics: Local Search and Optimization Metropolis Algorithm Greedy Optimization Hopfield Networks Max Cut Problem Nash

CS 781 Lecture 9 March 10, 2011 Topics: Local Search and Optimization Metropolis Algorithm Greedy Optimization Hopfield Networks Max Cut Problem Nash CS 781 Lecture 9 March 10, 2011 Topics: Local Search and Optimization Metropolis Algorithm Greedy Optimization Hopfield Networks Max Cut Problem Nash Equilibrium Price of Stability Coping With NP-Hardness

More information

CMPUT 675: Approximation Algorithms Fall 2014

CMPUT 675: Approximation Algorithms Fall 2014 CMPUT 675: Approximation Algorithms Fall 204 Lecture 25 (Nov 3 & 5): Group Steiner Tree Lecturer: Zachary Friggstad Scribe: Zachary Friggstad 25. Group Steiner Tree In this problem, we are given a graph

More information

Probability Models of Information Exchange on Networks Lecture 5

Probability Models of Information Exchange on Networks Lecture 5 Probability Models of Information Exchange on Networks Lecture 5 Elchanan Mossel (UC Berkeley) July 25, 2013 1 / 22 Informational Framework Each agent receives a private signal X i which depends on S.

More information

Hamiltonian paths in tournaments A generalization of sorting DM19 notes fall 2006

Hamiltonian paths in tournaments A generalization of sorting DM19 notes fall 2006 Hamiltonian paths in tournaments A generalization of sorting DM9 notes fall 2006 Jørgen Bang-Jensen Imada, SDU 30. august 2006 Introduction and motivation Tournaments which we will define mathematically

More information

The Algorithmic Foundations of Adaptive Data Analysis November, Lecture The Multiplicative Weights Algorithm

The Algorithmic Foundations of Adaptive Data Analysis November, Lecture The Multiplicative Weights Algorithm he Algorithmic Foundations of Adaptive Data Analysis November, 207 Lecture 5-6 Lecturer: Aaron Roth Scribe: Aaron Roth he Multiplicative Weights Algorithm In this lecture, we define and analyze a classic,

More information

MATH 56A SPRING 2008 STOCHASTIC PROCESSES

MATH 56A SPRING 2008 STOCHASTIC PROCESSES MATH 56A SPRING 008 STOCHASTIC PROCESSES KIYOSHI IGUSA Contents 4. Optimal Stopping Time 95 4.1. Definitions 95 4.. The basic problem 95 4.3. Solutions to basic problem 97 4.4. Cost functions 101 4.5.

More information

Notes on Blackwell s Comparison of Experiments Tilman Börgers, June 29, 2009

Notes on Blackwell s Comparison of Experiments Tilman Börgers, June 29, 2009 Notes on Blackwell s Comparison of Experiments Tilman Börgers, June 29, 2009 These notes are based on Chapter 12 of David Blackwell and M. A.Girshick, Theory of Games and Statistical Decisions, John Wiley

More information

Learning Methods for Online Prediction Problems. Peter Bartlett Statistics and EECS UC Berkeley

Learning Methods for Online Prediction Problems. Peter Bartlett Statistics and EECS UC Berkeley Learning Methods for Online Prediction Problems Peter Bartlett Statistics and EECS UC Berkeley Course Synopsis A finite comparison class: A = {1,..., m}. Converting online to batch. Online convex optimization.

More information

Resource Allocation via the Median Rule: Theory and Simulations in the Discrete Case

Resource Allocation via the Median Rule: Theory and Simulations in the Discrete Case Resource Allocation via the Median Rule: Theory and Simulations in the Discrete Case Klaus Nehring Clemens Puppe January 2017 **** Preliminary Version ***** Not to be quoted without permission from the

More information

1 Investing and combining expert advice

1 Investing and combining expert advice Cornell University, Fall 2013 CS 6820: Algorithms Lecture notes: The multiplicative weights update method October 9 11, 2013 The multiplicative weights update method is a family of algorithms that have

More information

Massachusetts Institute of Technology 6.854J/18.415J: Advanced Algorithms Friday, March 18, 2016 Ankur Moitra. Problem Set 6

Massachusetts Institute of Technology 6.854J/18.415J: Advanced Algorithms Friday, March 18, 2016 Ankur Moitra. Problem Set 6 Massachusetts Institute of Technology 6.854J/18.415J: Advanced Algorithms Friday, March 18, 2016 Ankur Moitra Problem Set 6 Due: Wednesday, April 6, 2016 7 pm Dropbox Outside Stata G5 Collaboration policy:

More information

Lecture Notes on Bargaining

Lecture Notes on Bargaining Lecture Notes on Bargaining Levent Koçkesen 1 Axiomatic Bargaining and Nash Solution 1.1 Preliminaries The axiomatic theory of bargaining originated in a fundamental paper by Nash (1950, Econometrica).

More information

A Fictitious Play Approach to Large-Scale Optimization

A Fictitious Play Approach to Large-Scale Optimization A Fictitious Play Approach to Large-Scale Optimization Theodore J. Lambert III Marina A. Epelman Robert L. Smith Department of Industrial and Operations Engineering University of Michigan, Ann Arbor, MI

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

Alg II Analyzing Functions ~1~ NJCTL.org. Domain and Range Class Work Find the domain and range for each of the following

Alg II Analyzing Functions ~1~ NJCTL.org. Domain and Range Class Work Find the domain and range for each of the following Domain and Range Class Work Find the domain and range for each of the following. 1. {(1,), (3,4), (5,6)}. {(4,3), (3,), (4,)} 3. {(5,1), (3,1), ( 4,1)} 4. 5. 6. 7. 8. 9. 10. 11. 1. 13. Domain and Range

More information

Approximation Basics

Approximation Basics Approximation Basics, Concepts, and Examples Xiaofeng Gao Department of Computer Science and Engineering Shanghai Jiao Tong University, P.R.China Fall 2012 Special thanks is given to Dr. Guoqiang Li for

More information

Game Theory, Alive. Anna R. Karlin and Yuval Peres. Two chapters from draft of upcoming book May 14, 2015

Game Theory, Alive. Anna R. Karlin and Yuval Peres. Two chapters from draft of upcoming book May 14, 2015 Game Theory, Alive Anna R. Karlin and Yuval Peres Two chapters from draft of upcoming book May 14, 2015 Please send comments and corrections to karlin@cs.washington.edu 1 Two-person zero-sum games We

More information

Chapter 1: Equations and Inequalities Section 1.1: Graphs and Graphing Utilities The Rectangular Coordinate System and Graphs

Chapter 1: Equations and Inequalities Section 1.1: Graphs and Graphing Utilities The Rectangular Coordinate System and Graphs Chapter 1: Equations and Inequalities Section 1.1: Graphs and Graphing Utilities The Rectangular Coordinate System and Graphs Coordinate Geometry y II I P(a, b) (a, b) = (x, y) O x III IV A relationship

More information

PERFECT SECRECY AND ADVERSARIAL INDISTINGUISHABILITY

PERFECT SECRECY AND ADVERSARIAL INDISTINGUISHABILITY PERFECT SECRECY AND ADVERSARIAL INDISTINGUISHABILITY BURTON ROSENBERG UNIVERSITY OF MIAMI Contents 1. Perfect Secrecy 1 1.1. A Perfectly Secret Cipher 2 1.2. Odds Ratio and Bias 3 1.3. Conditions for Perfect

More information

Pure Nash Equilibrium in A Capacitated Resource Allocation Game*

Pure Nash Equilibrium in A Capacitated Resource Allocation Game* Pure Nash Equilibrium in A Capacitated Resource Allocation Game* Seyed Rasoul Etesami, Tamer Başar Coordinated Science Laboratory, University of Illinois at Urbana-Champaign, arxiv:1404.3442v3 [cs.gt]

More information

Game Theory DR. ÖZGÜR GÜRERK UNIVERSITY OF ERFURT WINTER TERM 2012/13. Strict and nonstrict equilibria

Game Theory DR. ÖZGÜR GÜRERK UNIVERSITY OF ERFURT WINTER TERM 2012/13. Strict and nonstrict equilibria Game Theory 2. Strategic Games contd. DR. ÖZGÜR GÜRERK UNIVERSITY OF ERFURT WINTER TERM 2012/13 Strict and nonstrict equilibria In the examples we have seen so far: A unilaterally deviation from Nash equilibrium

More information

CS151 Complexity Theory. Lecture 14 May 17, 2017

CS151 Complexity Theory. Lecture 14 May 17, 2017 CS151 Complexity Theory Lecture 14 May 17, 2017 IP = PSPACE Theorem: (Shamir) IP = PSPACE Note: IP PSPACE enumerate all possible interactions, explicitly calculate acceptance probability interaction extremely

More information

MAT 111 Final Exam Fall 2013 Name: If solving graphically, sketch a graph and label the solution.

MAT 111 Final Exam Fall 2013 Name: If solving graphically, sketch a graph and label the solution. MAT 111 Final Exam Fall 2013 Name: Show all work on test to receive credit. Draw a box around your answer. If solving algebraically, show all steps. If solving graphically, sketch a graph and label the

More information

IB Math Standard Level Year 1: Final Exam Review Alei - Desert Academy

IB Math Standard Level Year 1: Final Exam Review Alei - Desert Academy IB Math Standard Level Year : Final Exam Review Alei - Desert Academy 0- Standard Level Year Final Exam Review Name: Date: Class: You may not use a calculator on problems #- of this review.. Consider the

More information

Area I: Contract Theory Question (Econ 206)

Area I: Contract Theory Question (Econ 206) Theory Field Exam Summer 2011 Instructions You must complete two of the four areas (the areas being (I) contract theory, (II) game theory A, (III) game theory B, and (IV) psychology & economics). Be sure

More information

Geometric Distribution The characteristics of a geometric experiment are: 1. There are one or more Bernoulli trials with all failures except the last

Geometric Distribution The characteristics of a geometric experiment are: 1. There are one or more Bernoulli trials with all failures except the last Geometric Distribution The characteristics of a geometric experiment are: 1. There are one or more Bernoulli trials with all failures except the last one, which is a success. In other words, you keep repeating

More information

Handout 1: Introduction to Dynamic Programming. 1 Dynamic Programming: Introduction and Examples

Handout 1: Introduction to Dynamic Programming. 1 Dynamic Programming: Introduction and Examples SEEM 3470: Dynamic Optimization and Applications 2013 14 Second Term Handout 1: Introduction to Dynamic Programming Instructor: Shiqian Ma January 6, 2014 Suggested Reading: Sections 1.1 1.5 of Chapter

More information

T 1. The value function v(x) is the expected net gain when using the optimal stopping time starting at state x:

T 1. The value function v(x) is the expected net gain when using the optimal stopping time starting at state x: 108 OPTIMAL STOPPING TIME 4.4. Cost functions. The cost function g(x) gives the price you must pay to continue from state x. If T is your stopping time then X T is your stopping state and f(x T ) is your

More information

Motivating examples Introduction to algorithms Simplex algorithm. On a particular example General algorithm. Duality An application to game theory

Motivating examples Introduction to algorithms Simplex algorithm. On a particular example General algorithm. Duality An application to game theory Instructor: Shengyu Zhang 1 LP Motivating examples Introduction to algorithms Simplex algorithm On a particular example General algorithm Duality An application to game theory 2 Example 1: profit maximization

More information

Microeconomic Theory (501b) Problem Set 10. Auctions and Moral Hazard Suggested Solution: Tibor Heumann

Microeconomic Theory (501b) Problem Set 10. Auctions and Moral Hazard Suggested Solution: Tibor Heumann Dirk Bergemann Department of Economics Yale University Microeconomic Theory (50b) Problem Set 0. Auctions and Moral Hazard Suggested Solution: Tibor Heumann 4/5/4 This problem set is due on Tuesday, 4//4..

More information

CS261: Problem Set #3

CS261: Problem Set #3 CS261: Problem Set #3 Due by 11:59 PM on Tuesday, February 23, 2016 Instructions: (1) Form a group of 1-3 students. You should turn in only one write-up for your entire group. (2) Submission instructions:

More information

Economics 2010c: Lecture 2 Iterative Methods in Dynamic Programming

Economics 2010c: Lecture 2 Iterative Methods in Dynamic Programming Economics 2010c: Lecture 2 Iterative Methods in Dynamic Programming David Laibson 9/04/2014 Outline: 1. Functional operators 2. Iterative solutions for the Bellman Equation 3. Contraction Mapping Theorem

More information

Empirical Risk Minimization Algorithms

Empirical Risk Minimization Algorithms Empirical Risk Minimization Algorithms Tirgul 2 Part I November 2016 Reminder Domain set, X : the set of objects that we wish to label. Label set, Y : the set of possible labels. A prediction rule, h:

More information

Equilibrium Refinements

Equilibrium Refinements Equilibrium Refinements Mihai Manea MIT Sequential Equilibrium In many games information is imperfect and the only subgame is the original game... subgame perfect equilibrium = Nash equilibrium Play starting

More information

Game Theory. Greg Plaxton Theory in Programming Practice, Spring 2004 Department of Computer Science University of Texas at Austin

Game Theory. Greg Plaxton Theory in Programming Practice, Spring 2004 Department of Computer Science University of Texas at Austin Game Theory Greg Plaxton Theory in Programming Practice, Spring 2004 Department of Computer Science University of Texas at Austin Bimatrix Games We are given two real m n matrices A = (a ij ), B = (b ij

More information

Practice Final Exam Answers

Practice Final Exam Answers Practice Final Exam Answers 1. AutoTime, a manufacturer of electronic digital timers, has a monthly fixed cost of $48,000 and a production cost $8 per timer. The timers sell for $14 apiece. (a) (3 pts)

More information

DRAFT Formulation and Analysis of Linear Programs

DRAFT Formulation and Analysis of Linear Programs DRAFT Formulation and Analysis of Linear Programs Benjamin Van Roy and Kahn Mason c Benjamin Van Roy and Kahn Mason September 26, 2005 1 2 Contents 1 Introduction 7 1.1 Linear Algebra..........................

More information

5.2 A characterization of the nonemptiness of the core

5.2 A characterization of the nonemptiness of the core Computational Aspects of Game Theory Bertinoro Spring School Lecturer: Bruno Codenotti Lecture 5: The Core of Cooperative Games The core is by far the cooperative solution concept most used in Economic

More information

Economics 703 Advanced Microeconomics. Professor Peter Cramton Fall 2017

Economics 703 Advanced Microeconomics. Professor Peter Cramton Fall 2017 Economics 703 Advanced Microeconomics Professor Peter Cramton Fall 2017 1 Outline Introduction Syllabus Web demonstration Examples 2 About Me: Peter Cramton B.S. Engineering, Cornell University Ph.D. Business

More information

A Polynomial-Time Algorithm for Pliable Index Coding

A Polynomial-Time Algorithm for Pliable Index Coding 1 A Polynomial-Time Algorithm for Pliable Index Coding Linqi Song and Christina Fragouli arxiv:1610.06845v [cs.it] 9 Aug 017 Abstract In pliable index coding, we consider a server with m messages and n

More information

Multiobjective Mixed-Integer Stackelberg Games

Multiobjective Mixed-Integer Stackelberg Games Solving the Multiobjective Mixed-Integer SCOTT DENEGRE TED RALPHS ISE Department COR@L Lab Lehigh University tkralphs@lehigh.edu EURO XXI, Reykjavic, Iceland July 3, 2006 Outline Solving the 1 General

More information