# Register now to access 7 million high quality study materials (What's Course Hero?) Course Hero is the premier provider of high quality online educational resources. With millions of study documents, online tutors, digital flashcards and free courseware, Course Hero is helping students learn more efficiently and effectively. Whether you're interested in exploring new subjects or mastering key topics for your next exam, Course Hero has the tools you need to achieve your goals.

1 Page

### lecture14

Course: CS 4141, Fall 2009
School: Allan Hancock College
Rating:

Word Count: 829

#### Document Preview

Algorithms COMP4141 Games Probabilistic Theory of Computation Lecture 14 Games Kai Engelhardt CSE, UNSW (and NICTA) Revision: 1.2 of Date: 2007/02/01 08:51:09 UTC (Credits: D Dill, R v Glabbeek, M Sipser, W Thomas, T Wilke) 1 Games Probabilistic Algorithms Games Game theory nowadays plays an important role in economics, politics, and computer science. Lets consider two-player turn-based games such as...

Register Now

#### Unformatted Document Excerpt

Coursehero >> California >> Allan Hancock College >> CS 4141

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
Algorithms COMP4141 Games Probabilistic Theory of Computation Lecture 14 Games Kai Engelhardt CSE, UNSW (and NICTA) Revision: 1.2 of Date: 2007/02/01 08:51:09 UTC (Credits: D Dill, R v Glabbeek, M Sipser, W Thomas, T Wilke) 1 Games Probabilistic Algorithms Games Game theory nowadays plays an important role in economics, politics, and computer science. Lets consider two-player turn-based games such as Tic-Tac-Toe, Checkers, Chess, Go, Geography Game, etc. only. A position is a comprehensive description of the game state. Example is such a description for Tic-Tac-Toe (assuming that the player placing always moves rst). Histories of moves are always usable as positions, however, they may be not concise enough for practical purposes. 2 Games Probabilistic Algorithms Graphs and Winning Strategies In general, the possible positions of (nite) games form directed graphs where the edges are sometimes labelled with the player who can move from the position at the source of the edge to the position at the destination. When we use histories as positions then graphs become trees, odd level moves belong to one player whereas even level moves belong to the other. The leaves in such a tree describe nal positions of the game. These leaves are marked depending on who wins the game when it ends in this position. A player has a winning strategy in a position if, by choosing only his moves from that position onwards, he can force a win, regardless of the other players moves. 3 Games Probabilistic Algorithms Connection to Logic The question does player E have a winning strategy in position p of game G can often be re-phrased as a quantied Boolean formula m1 (r1 (m2 (. . . ))) where the mi and ri range over player E s, resp. As moves and expresses that E wins. 4 Games Probabilistic Algorithms Formula Game Lets consider a simple formula game. Let be a Boolean formula over propositions x1 , . . . , xk . In move i a player chooses a truth value for xi . Player E takes the odd turns and player A takes the even turns. Player E wins if the formula turns out true, otherwise A wins. Player E has a winning strategy i the quantied Boolean formula x1 (x2 (xr (. . . ))) is true. 5 Games Probabilistic Algorithms Formula Game cont. Since it is pretty much the same as QBF it follows that { | E has a winning strategy for } is also PSPACE-complete. Sipser puts the quantiers into but this doesnt a make dierence because we assume an order on the propositions and from that the quantier sequence follows. 6 Games Probabilistic Algorithms Geography Sipser writes that kids play a game in which one starts with a city name and the next kid always has to come up with a new city name that starts with the last letter of the previous one. The kid who cannot name another city in this sequence loses. We could re-phrase this as a graph problem by making all the known city names nodes and have edges from a name wa to bv i a = b. Or we could generalise this even further to an arbitrary directed graph and forget about the city names. GG = { G , b | Kid 1 has a w. s. in G starting at b } 7 Games Probabilistic Algorithms Geography cont. Theorem GG is PSPACE-complete. Proof of GG PSPACE. That GG is in PSPACE follows from having a polynomial space TM M which decides GG. On input (V , E ), b : 1 Reject if b has no successor nodes in (V , E ) because kid 1 loses. Remove b from (V , E ): let V1 = V \ {b} and E1 = E (V1 V1 ). For each b1 such that (b, b1 ) E run M on (V1 , E1 ), b1 . If all of these recursive calls accept, then kid 2 has a winning strategy in the game so reject, otherwise accept. 2 3 4 8 Games Probabilistic Algorithms Geography cont. Proof of GG is PSPACE-hard. by reduction from QBF. 9 Games Probabilistic Algorithms Real Games Real games such as Chess and Go have xed board sizes and thus nite numbers of p...

Textbooks related to the document above:
Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

Purdue - ME - 576
ME 576 Homework #5 1. 2. Bollinger 4.3 Bollinger 12.1Due: 4/2/2009Name :(Hint: the discrete root corresponding to time constant is e 3.TDesign a discrete proportional-derivative (PD) controller for the control system shown below such tha
Idaho - CH - 310
Idaho - CH - 310
Auburn - CTMU - 7510
Chapter 2Quantitative and Qualitative ApproachesEducational Research by John W. Creswell. Copyright 2002 by Pearson Education. All rights reserved.Slide 1Key Ideas4Configurations of research 4Development of quantitative research 4Development
Vassar - CS - 101
CMPU-101-02, Fall 2008Asmt 7Due: Monday, Nov. 17 @ 1:30 p.m.For this assignment, you will design a more complete Yahtzee gameusing object-oriented programming. A big difference will be that youwill take care of scoring. In addition, you wil
Vassar - CS - 295
; -; CMPU-295, Spring 2009; Code used in class; Apr. 1; -; Propositional Logic in Scheme; We'll represent sentences in PROPOSITIONAL LOGIC; using SYMBOLS for propositional letters and; LISTS of the form (NOT S), (S1 AND S2), (
Vassar - CS - 295
; -; CMPU-295, Spring 2009; Code used in class; Apr. 6, 2009; -; This file contains the stuff on Propositional Logic; introduced on Apr. 1, as well as the additional stuff; we did in class on Apr. 6; -; Recall:; We represent s
Vassar - CS - 245
; =; CMPU-245, Spring 2009; Code used in class; =; Implementing Arithmetic in the Lambda Calculus; First, a note about Scheme and &quot;currying&quot; of inputs.; FUNC takes two inputs at once.; (e.g., (func 3 4) )(define func (lambda (
Purdue - MA - 315
MA 315, Spring 2009 Homework 11 Due Tuesday, April 14, 2009 Prove the following. 1. Let f : X Y be a function. Assume A and B are subsets of Y . Prove or disprove: f -1 (A B) = f -1 (A) f -1 (B).2. Let f : X Y . For each element b Y , let Yb =
Auburn - DIGDATA - 1903
149REPORT OF THE PRESIDENTto theTrustees of the Alabama Polytechnic Institute, June 8, 1903 The friends of the college and of education have cause for gratification at the continued growth and prosperity of the institution, a prosperity the
Idaho - CH - 310
Auburn - AG - 1987
Conservation Tillage: Today and Tomorrow Southern Region No-till ConferenceProceedingsJuly, 1987 College Station, Texas The Texas Agricultural Experiment Station, Neville P. Clarke,Director, The Texas A&amp;M University System, College Statio
Auburn - AG - 1990
1990 Southern Region Conservation Tillage Conference Steering Committee J. S. Bacheler, Department of Entomology, NCSU C. G. Bowers, Department of Biological and Agricultural Engineering, NCSU B. G. Brock, Soil Conservation Service M. G. Co
Auburn - AG - 1985
169 Effects of Tillage on Quality of Runoff WaterP. L. Baldwin, W. W. Frye, and R. L. BlevinsDepartment of Agronomy, University of Kentucky, Lexington, KY 40546The i s s u e of water q u a l i t y i s o f g r e a t importance t o a g r i c u l
Auburn - AG - 1987
PrefaceWe recognize the importance of tillage in modern agriculture especially in respect to ameliorating biological, chemical, and physical soil impediments to crop growth. Modern tillage practices have contributed to the unmatched productivity of
Allan Hancock College - CS - 9311
Allan Hancock College - CS - 9311
SQL: QueriesexampleDB-1SQL Query LanguageAn SQL query consists of a sequence of clauses: SELECT FROM WHERE GROUP BY HAVING projectionList relations/joins condition groupingAttributes groupConditionWHERE, GROUP BY, HAVING clauses are optional.
Allan Hancock College - CS - 9311
EII DB Day 2007Database Group @ UNSW, Australia 2 October 2007Sponsered by ARC Research Network in Enterprise Information Infrastructure (EII)Database Group School of Computer Science and Engineering University of New South Wales Sydney 2052, NS
Vassar - CS - 240
Reducing One Undecidable Problem to Another We say a probem A is reduced to problem B is the decidability of A follows from the decidability of B Then, if we know A is undecidable, we can conclude that B is also undecidable The typical case is to
Allan Hancock College - CS - 9311
Data Modification in SQLSQL provides mechanisms for modifying data (tuples) in tables: INSERT . add a new tuple into a tableSQL: Updates DELETE . remove tuples from a table (via condition) UPDATE . modify values in exiting tuples (via conditio
Allan Hancock College - CS - 9311
RDBMS Architecture What is an RDBMS?A relational database management system (RDBMS) is software designed to support large-scale data-intensive applications allowing high-level description of data (domains, constraints) with high-level access to t
Allan Hancock College - CS - 9311
Data Modication in SQLSQL provides mechanisms for modifying data (tuples) in tables: INSERT . add a new tuple into a tableSQL: Updates DELETE . remove tuples from a table (via condition) UPDATE . modify values in exiting tuples (via condition)
Allan Hancock College - CS - 9311
Future of DatabaseCore &quot;database&quot; goals: deal with very large amounts of data (terabytes, petabyes, .) very-high-level languages (deal with big data in uniform ways) query optimisation(evaluation too slow useless)Future of Database(cont.)
Allan Hancock College - CS - 9311
RDBMS Architecture What is an RDBMS?A relational database management system (RDBMS) is software designed to support large-scale data-intensive applications allowing high-level description of data (domains, constraints) with high-level access to t
Allan Hancock College - CS - 9311
Data ModellingData modelling: an important early stage of database application development (aka &quot;database engineering&quot;). 1. requirements analysis 2. data modelling(identify data and operations)(high-level, abstract) (detailed, relational/tables)
Purdue - CS - 251
z y o &amp;X' VB y 8' c I d 2 02 2 z y {# o &amp;X' VB VB cvrfHgs I j I d ctp n rp n j d 3# o s s q p 2 nl j VB I ' Tv' cvg&amp; Tv{v8 v j fC o d y ' } c d c H U fu &amp; vy Qo U c P d d y # 8 n c&amp; fd BI r QY vgd 2 0~Tv' c D s y 8 ' 2 ' c y 8'2 y'
Purdue - CS - 514
CS514 : Numerical AnalysisHomework 4 Solution Problem 1 The attached Matlab computes f(0) and f(0) using divided differences and origin shifts. The values were found to be f(0) = 0.6788 f(0) = -0.2876 Problem 2 We approximate the step function f(t
Purdue - MA - 490
MA49000 HOMEWORKASSIGNMENT 8: SELECTED PROBLEM SOLUTIONSProblem 4, Ch. 5. If g is a smooth function on R, define the formal power series(1)u(x, t) =n=0g (n) (t)x2n . (2n)!(a) Check formally that u solves the heat equation. (b) For a &gt;
Vassar - CS - 335
Using Design Patterns to Elaborate upon Design ModelsMoving from Analysis to DesignExamine Design models and initial code to:Improve cohesion Reduce coupling Enhance ReusabilityGoF Design Patterns should be used to rewrite code to promote these
RPI - MB - 2400
RPI - MATH - 2400
RPI - MATH - 2400
Purdue - STAT - 416
Name Student ID # SectionSOLUTIONSTAT/MATH 416 Spring 09 Practice Midterm #2 April 9, 2009You are not allowed to use books or notes. Calculators approved for the actuarial exams are permitted. The last page contains a table for the cumulative di
Purdue - STAT - 416
STAT 416 Spring 09 Section 001 Homework 10 Due: April 16, 2009, 9:00amApril 8, 2009Please explain how you arrived at your solution, don't just write down the final number. Reading: Chapter 6.5,6.7, 7.1-7.2 Exercises: problems 6.51, 6.52, 6.54, 6.
RPI - KU - 4800
RPI - KU - 4800
RPI - KU - 4800
RPI - KU - 4800
RPI - KU - 4800
RPI - ECSE - 35201
35201 SAMPLE FINAL EXAM 5/6/97 This exam has two parts. There are 8 problems weighted as shown. Put your answers in the spaces provided. Include units where appropriate. SHOW YOUR WORK! Please print your name on EVERY page, including any attachments.
RPI - ECSE - 35201
35201 FINAL EXAM Take-Home Part (Problems 1-4) In-Class Part (Problems 5-8)SAMPLE EXAMPROB. 1 (15 points) i 10 mA 0.4 F 20 k 5 k t=0 15 k i(0-) = 8 mA, i(0+) = 2 mA, Iss = 5 mA, = 4 ms PROB. 2 (10 points) Given H ( s) = , plot and label the pole
TCNJ - CSC - 220
Csc220 p. 195 2.Self-check answerspart 2a. a = (b + a b) 3rd 1st 2nd b. (c = (a + b) | flag 2nd 1st 3rd c. (a != 7) &amp; (c &gt;= 6) | flag 1st 3rd 2nd 4th d. !(b &lt;= 12) &amp; (a %2 = 0) 4th 1st 5th 2nd 3rd e. !(a &gt; 5) | (c &lt; (a + b) 5th 1st 4th 3rd 2nd
Purdue - STAT - 598
rnorm.ar = function (n) {# generating a sample of size n from normal N(0, 1) using the # acceptance-rejection method. x = rep(0, n) R = sqrt(exp(1)/2/pi) for(i in 1:n){ while(TRUE){ # Step 1. Y = -log(runif(1)
RPI - P - 0591
ANNEXWPP: A Dynamic decision making taskIn WPP the decision maker plays the role of an operator of a Water Purification Plant, whose goal is to distribute water to different locations on time. To do the task, the decision maker activates/de-activa
Auburn - DIGDATA - 1867
72ON MOTION, RESOLVED, That a committee of three be appointed to make an appealto the public in behalf of the College and also to present to the Montgomery District meeting in September the present status of the College and its claims upon public
UNLV - ECG - 495
1. open up simplescalar.xla file2. press simpledata toolbar button in excel3. choose directory in which all benchmark output &quot;.txt&quot; files are located4. choose which values you want the program to grab (separated by commas).Example: &quot;sim_num_i
UNLV - CS - 370
Operating Systems CS 370 Spring 2006 Project #1 Due: Thursday, February 2, 2006 DOS to Linux Command Shell Interpreter Total points: 100Teaching Assistants: Meghna Sharma and Ai Yamazaki Office Hours in TBE B-361: MW - 10:30 - 11:30 AM and 2 - 3 PM
UNLV - CS - 472
CS472 Software Product Design ISpring 2008 Topic PresentationObjectives: Students will practice oral communication skills Students will practice written communication skills Students will specialize in a software engineering topic and learn addit
RPI - COMM - 6820
Maureen Duffy &lt;duffy@rpi.edu&gt; COMM 6820: Foundations of HCI Usability Assignment #2 Interface Metaphors AN EVALUATION OF WINAMP'S STEREO COMPONENT METAPHOR INTRODUCTION Winamp 3 is a free media player downloadable from winamp.com. It was originally
TCNJ - SPED - 506
VISION STATEMENT:CommunicationWe are committed to insuring that our technology effectively supports and encourages communication between teachers, students, parents, and administrators to establish a greater sense of community and ownership in the
Centenary LA - REL - 320
Virtual Tour of Ancie Myce nt naePhotographs by Margare t-AnneGillis, BarrieCe ntral Colle giate , Barrie Ontario , and Elizabe Ellison, th Elm wood S chool Ottawa, Ontario CANADA1Myce , se n fromthevalle surrounding it. nae e yPhoto: Elizabe E
TCNJ - PHY - 466
TCNJ - PHY - 466
APPENDIX A1. Trigonometric DistancesThe distance of an object may be found by measuring an angle called parallax. Parallax is defined to be the angular displacement of an object when viewed from two different positions that are located along a line
TCNJ - PHY - 416
CHAPTER 44.1 Work We define work to be the interchange of energy between a system and it surroundings under the action of a force. In general: = where is the angle between the force vector and the direction of the displacement ds. In thermodynamics
TCNJ - AST - 161
ASTR161 TEST 2, March 4, 2009Part I. Multiple Choice01. 02. 03. 04. 05. 06. 07. 08. 09. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. B E M V J N D C E H V A B V J D B B A D B A C C D A A D D E B C B E B A28. 29. 30. 31
TCNJ - PHY - 416
D06-2, 6-3, 6-4a, 6-17Section 6-9. Properties of a solid or liquid under hydrostatic pressProblem 6-22 in S&amp;S was done as an example in class. Do 6-22D.
TCNJ - PHY - 466
CHAPTER 1 THE OBSERVATIONAL PROPERTIES OF STARS1-1. PhotometryPhotometry involves the measurement of the properties of light or electromagnetic radiation in general. More specifically, it deals with measuring the brightness of a light source, wheth
TCNJ - PHY - 466
CHAPTER 4B4-4 OpacityWe now turn to the matter of determining what contributes to the optical depth of a medium, such as the layers of a star. The theory here can be quite complex and a thorough exposition would take us deep into quantum mechanics.
Auburn - MECH - 4240
Gang Saw GangMohammad Jeelani Richard Thrift Barton Pate Michael Knoblett Jeremy Patterson Mark Herndon Justin Owens Joseph Perry Mike HudgensProblemTo design a reliable machine to cut various sizes of scrap wood into survey stakes of specific di
Auburn - ELEC - 6740
ELEC 6740: Electronics Manufacturing Chapter 12: Soldering of Surface Mount ComponentsR. Wayne Johnson Alumni Professor ECE Department, Auburn University 334-844-1880 johnson@eng.auburn.eduSubjects! ! ! ! ! ! !Introduction Wave Soldering Vapor