Constraint Games. Huawei Technologies Ltd University of Caen-Normandie. Arnaud Lallouet joint work with Anthony Palmieri GDR IA-RO

Size: px
Start display at page:

Download "Constraint Games. Huawei Technologies Ltd University of Caen-Normandie. Arnaud Lallouet joint work with Anthony Palmieri GDR IA-RO"

Transcription

1 Constraint Games Huawei Technologies Ltd University of Caen-Normandie GDR IA-RO Arnaud Lallouet joint work with Anthony Palmieri

2 Huawei Technologies Co., Ltd. 2 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

3 Huawei Technologies Co., Ltd. 3 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

4 Huawei Technologies Co., Ltd. 4 Huawei Technologies Ltd Huawei s vision: Building a better connected world

5 Artificial Intelligence in Huawei Huawei Technologies Co., Ltd. 5

6 Artificial Intelligence in Huawei Huawei Technologies Co., Ltd. 6

7 French Research Center Huawei Technologies Co., Ltd. 7

8 Huawei Technologies Co., Ltd. 8 Artificial Intelligence in Huawei Research done in FRC Smartphone (see new Kirin 970: CPU+GPU+NPU) Cloud services Network applications, SDN, content delivery Internal use: recommendation, big data analytics 5G for connected and autonomous car and IOT An integration strategy Mostly integration of software Contributions to open source AI in Transition from Hardware to Software

9 Huawei Technologies Co., Ltd. 9 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

10 Huawei Technologies Co., Ltd. 1 Strategic games: the setting Set of Players P Each player i performs Actions A i and wants to maximize an Utility depending on other players actions Each player i has its own utility function (u i ), or preference Strategic form, also called multimatrix model In this talk: finite games only (also simultaneous, perfect information, selfish players)

11 Huawei Technologies Co., Ltd. 1 Solution concept Deviation We say that a player has a beneficial deviation if he can increase its utility by changing its own strategy How can we tell when a player is satisfied? Find a point where each player chooses the best strategy for him/herself... and for which no player can improve his/her utility by changing to another action: Pure Nash Equilibrium

12 A generic algorithm to solve games Solve function solve(s): tuple for s ΠA do if nash(s) then return s end if end for return not found Deviation function deviation(s, i): boolean for v A i, v s i do if u i (v.s i ) > u i (s) then return true end if end for return false Nash function nash(s): boolean for i P do if deviation(s, i) then return false end if end for return true Analysis Inefficient but still the baseline algorithm Implemented in the Gambit solver along with IESDS [McKelvey and al, 2010] Huawei Technologies Co., Ltd. 1

13 Game representation Matrix in Normal Form games grow exponentially with the number of players Exploiting Structure: Graphical Games[KLS01] dependencies between player are provided with a graph to reduce the matrix size Exploiting Semantics: Action-graph Games [XKNB07] Symmetry + anonymity Exploiting Semantics: Boolean Games [HvdHMW01] Represents utilities with SAT problems Like SAT for satisfiability, Boolean Games are a fundamental tool to study interaction between agents Huawei Technologies Co., Ltd. 1

14 Huawei Technologies Co., Ltd. 1 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

15 Constraint Games The idea Use CSP to express utilities Constraint Satisfaction Game A Constraint Satisfaction Game (CSG) is a 4-uple CG = (P, V, D, G) where P is a set of players V is a set of variables, Player i controls V i V D = (D x ) x V defines a (finite) domain for each variable G = (G i ) i P is a family of CSP Preferences CSPs provide a compact and natural formalism to express satisfaction for a player: G i is called Goal of Player i Goals express preferences and an equilibrium may hold if a player is not satisfied (and cannot be) Huawei Technologies Co., Ltd. 1

16 Huawei Technologies Co., Ltd. 1 COG and hard constraints Hard constraints CSG/COG can be enhanced with a set of hard constraints (HC) to forbid invalid equilibria a strategy profile which does not satisfy HC cannot be an equilibrium and should not be checked for deviations impossible to represent in the matrix model (even by giving a dummy value) Constraint Optimization Games Constraint Programming provides an easy way to express optimization: add min(x ) or max(x ) to the goal of each player Allows to represent in a natural way many useful games (see examples after)

17 Huawei Technologies Co., Ltd. 1 Example: Collaboration game Defintion two drivers meet on a narrow dirty road. Both have to swerve in order to avoid a head-on collision Right Left Left Right A simple model, is to choose an action, the utility of a player corresponds to the number of player choosing the same action

18 Collaboration game as Constraint Game p u b l i c c l a s s C o l l a b o r a t i o n G a m e extends AbstractGameModel { p r i v a t e f i n a l i n t n b S t r a t ; p u b l i c C o l l a b o r a t i o n G a m e ( i n t n b P l a y e r s, i n t n b S t r a t ) { s u p e r ( n b P l a y e r s ) ; t h i s. nbstrat = nbstrat ; O v e r r i d e p u b l i c v o i d b u i l d M o d e l ( Model m) { IntVar Choice [ ] = m. intvararray ( " Choice ", players. length 1, 0, nbstrat 1 ) ; IntVar Objectives [ ] = m. intvararray ( " objective ", players. length 1, 1, players. length ) ; f o r ( i n t i = 0 ; i < p l a y e r s. l e n g t h 1; i ++) { IntVar Evalue = m. intvar (m. generatename ( "COUNT_ " ), 0, players. length ) ; m. count ( Evalue, Choice, O b j e c t i v e s [ i ] ). p o s t ( ) ; ; } } f o r ( i n t i = 1 ; i < p l a y e r s. length ; i++) { p l a y e r s [ i ]. own ( Choice [ i 1 ] ) ; p l a y e r s [ i ]. s e t O b j e c t i v e ( R e s o l u t i o n P o l i c y.maximize, O b j e c t i v e s [ i 1 ] ) ; } } p u b l i c s t a t i c v o i d main ( S t r i n g [ ] a r g s ) { CongaSolver cg = new CongaSolver (new CollaborationGame ( 5, 5 ) ) ; cg. s e t C o n s t r a i n t B u i l d e r ( C o n s t r a i n t F a c t o r y.bound CONSTRAINT ) ; Algorithm a=a l g o r i t h m F a c t o r y.complete SEARCH. b u i l d ( cg ) ; a l g o. e x e c u t e ( ) ; } Huawei Technologies Co., Ltd. 1

19 ConGa: a solver for Constraint Games Fully integrated in CP solver Complete tree structured search Based on Choco solver V4 [PFL17] Nash constraint: global constraint to express player s preferences Different filtering power for Nash Constraints [NL14, PL17] Easy to extend Many available and reusable algorithms Price of anarchy Price of stability Pure Nash equilibria enumeration Pareto Nash equilibria Iterated Best Response search Huawei Technologies Co., Ltd. 1

20 Huawei Technologies Co., Ltd. 2 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

21 ConGa 1.0: A complete algorithm [Nguyen & Lallouet 2014] A result by [Gottlob and al, 2005] Nash Constraint N i for Player i encodes deviations, i.e. tuples t = (s i, s i ) such that s i is a best response to s i (not unique) Theorem: i P N i = PNE In Conga 1.0, we compute incrementally the N i First Tree-search algorithm The idea is to traverse all tuples of the search space using a complete ordering of players and values Record each player s undominated strategies Pruning when a tuple has already been proved subject to deviation (complete detection) Pruning when a tuple is NBR at last level Constraint solver is used to compute hard constraints and deviations Huawei Technologies Co., Ltd. 2

22 Huawei Technologies Co., Ltd. 2 ConGa 2.0: A new complete algorithm [PL17] Again using Nash constraints [Gottlob and al, 2005] In Conga 2.0, we filter the N i up to a certain level of consistency Tree-search algorithm Like in Conga 1.0, complete traversal of the search space Each player has an implicit Nash constraint to encapsulate preferences Nash constraint is implemented as a global constraint Arc-consistency is costly compute an approximation Better integration in the solver, better filtering... but no hard constraints

23 Nash constraint filtering: arc-consistency Arc-consistency for N i Nash constraint is of the same arity as the problem Arc-consistency consists in removing from D i values which do not belong to any best response Theorem Deciding if a is a NBR is Π P 2 -complete Proof Intuitively, check that for all assignments of the other players, there exists a better response than a Reduction from a QCSP to a 2-players 0-sum CSG Too complex to be implemented Huawei Technologies Co., Ltd. 2

24 Nash constraint filtering: approximations Approximation #1: filtering only objective values Decision variable values will be filtered from the expression of the objective function If a non best response state has same objective value than a best response state, its values will not be filtered Approximation #2: filtering only objective bounds Other inconsistent values are not filtered Computation of lower bound Mm(S) = max s i A i min s i S i s opti Approximation #3: use AC instead of traversal of S i to evaluate min s i S i s opti Complete search over A i (not S i because deviations are global for a player!) Branch and Bound on lower bound of the objective: operator BB i (S) Huawei Technologies Co., Ltd. 2

25 Huawei Technologies Co., Ltd. 2 Nash constraint filtering in practice Implementation of BB i Launch a new solver instance at choice point of the main search tree Start enumeration of all strategies for Player i to check deviations and remove NBR Theorem Main search tree BB i (S) is a propagator for N i Proof Check that BB is correct, contracting, monotonic and singleton complete Deviation search tree

26 Huawei Technologies Co., Ltd. 2 Special handling of Graphical Games Nash constraints in Graphical Games Use the dependency graph to restrict the arity of the Nash constraints Less wake-up conditions, only useful propagations Dynamic dependencies for free Nash checking for a player when its objective is instanciated Can appear in the middle of a search tree Entails the Nash constraint

27 Huawei Technologies Co., Ltd. 2 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

28 Benchmark suite Gamut: a library of game generators The games (only n-players) (#players,#strategies) MEG: Minimum Effort Game (payoff am be where M is the min effort. Coordination game) TD: Traveller s Dilemma DG: Dispersion Game (anti-coordination game, models load balancing) CG: Collaboration Game (the highest payoffs when all player choose the same action) AR: Arm Race (x and y are arm levels. Payoff is C(x) + B(x y). Symmetric) EFBG: El Farol Bar Game (If less than 60% of the population go to the bar, they ll all have a better time than if they stayed at home) CB: Colonel Blotto Game (distribution of resources along battlefields) (#players,#battlefields,#troops) Huawei Technologies Co., Ltd. 2

29 Huawei Technologies Co., Ltd. 2 The solvers Comparison of different solving techniques enum: this solver performs a complete enumeration of all the search space. This solver is similar to Gambit except that it uses the Constraint Game representation Settings Conga 1.0: this solver uses tables to record best responses and pruning at the last level Conga 2.0: this solver uses filtering on Nash constraints All times are given in seconds Time out is set to 600s

30 Experimental results Comparison of the different algorithms on Gamut games Game Param. #PNE enum Conga 1.0 Conga MEG TD DG CG AR EFBG CB Huawei Technologies Co., Ltd. 3

31 Experimental results Comparison of the different algorithms on graphical games Topology is C (circle, #players), T (tree, #players, #highest degree) and B (complete bipartite, #players) Game Topology #PNE enum Conga 1.0 Conga 2.0 C C T T PGG B B B C C C T TGC T B B B C C C RG T T B B B Huawei Technologies Co., Ltd. 3

32 Huawei Technologies Co., Ltd. 3 Contents 1 Huawei French Research Center 2 Game theory 3 Constraint Games 4 Solving methods 5 Benchmarks 6 Conclusion and perspectives

33 Huawei Technologies Co., Ltd. 3 Conclusion and perspectives Complete and generic solver Full integration in Choco: global constraints, heuristics, etc. Improvement by orders of magnitude for Pure Nash Equilibrium enumeration Symmetries: generalization of symmetry properties to Constraint games Heuristics (for first, and all solutions) Application to SDN

34 Huawei Technologies Co., Ltd. 3 Thank you for your attention Questions?

35 Bibliography I Paul Harrenstein, Wiebe van der Hoek, John-Jules Ch. Meyer, and Cees Witteveen. Boolean Games. In Johan van Benthem, editor, TARK. Morgan Kaufmann, Michael J. Kearns, Michael L. Littman, and Satinder P. Singh. Graphical models for game theory. In Jack S. Breese and Daphne Koller, editors, UAI, pages Morgan Kaufmann, Thi-Van-Anh Nguyen and Arnaud Lallouet. A complete solver for constraint games. In Barry O Sullivan, editor, Principles and Practice of Constraint Programming - 20th International Conference, CP Springer, Charles Prud homme, Jean-Guillaume Fages, and Xavier Lorca. Choco Documentation. TASC, INRIA Rennes, LINA CNRS UMR 6241, COSLING S.A.S., Huawei Technologies Co., Ltd. 3

36 Bibliography II Anthony Palmieri and Arnaud Lallouet. Constraint games revisited. In IJCAI, Albert Xin, Jiang Kevin, Leyton-brown Navin, and AR Bhat. Action-graph games

Constraint Games Revisited

Constraint Games Revisited Constraint Games Revisited Anthony Palmieri, Arnaud Lallouet Huawei Technologies Ltd, French Research Center Greyc, University of Caen - Normandie {anthony.palmieri,arnaud.lallouet}@huawei.com Abstract

More information

Computing Pure Nash Equilibria in Symmetric Action Graph Games

Computing Pure Nash Equilibria in Symmetric Action Graph Games Computing Pure Nash Equilibria in Symmetric Action Graph Games Albert Xin Jiang Kevin Leyton-Brown Department of Computer Science University of British Columbia {jiang;kevinlb}@cs.ubc.ca July 26, 2007

More information

Scaling Up Game Theory: Representation and Reasoning with Action Graph Games

Scaling Up Game Theory: Representation and Reasoning with Action Graph Games Scaling Up Game Theory: Representation and Reasoning with Action Graph Games Kevin Leyton-Brown Computer Science University of British Columbia This talk is primarily based on papers with: Albert Xin Jiang

More information

Bayesian Action-Graph Games

Bayesian Action-Graph Games Bayesian Action-Graph Games Albert Xin Jiang and Kevin Leyton-Brown Department of Computer Science University of British Columbia November 13, 2011 Equilibrium Computation in Bayesian Games Equilibrium

More information

Propagate the Right Thing: How Preferences Can Speed-Up Constraint Solving

Propagate the Right Thing: How Preferences Can Speed-Up Constraint Solving Propagate the Right Thing: How Preferences Can Speed-Up Constraint Solving Christian Bessiere Anais Fabre* LIRMM-CNRS (UMR 5506) 161, rue Ada F-34392 Montpellier Cedex 5 (bessiere,fabre}@lirmm.fr Ulrich

More information

Validating Plans with Durative Actions via Integrating Boolean and Numerical Constraints

Validating Plans with Durative Actions via Integrating Boolean and Numerical Constraints Validating Plans with Durative Actions via Integrating Boolean and Numerical Constraints Roman Barták Charles University in Prague, Faculty of Mathematics and Physics Institute for Theoretical Computer

More information

Pure Nash Equilibria: Complete Characterization of Hard and Easy Graphical Games

Pure Nash Equilibria: Complete Characterization of Hard and Easy Graphical Games Pure Nash Equilibria: Complete Characterization of Hard and Easy Graphical Games ABSTRACT Albert Xin Jiang Department of Computer Science University of British Columbia Vancouver, Canada jiang@cs.ubc.ca

More information

Learning Graphical Game Models

Learning Graphical Game Models Learning Graphical Game Models Quang Duong Yevgeniy Vorobeychik Satinder Singh Michael P. Wellman Computer Science & Engineering University of Michigan Ann Arbor, MI 4819-2121 USA Computer & Information

More information

Multiple Agents. Why can t we all just get along? (Rodney King) CS 3793/5233 Artificial Intelligence Multiple Agents 1

Multiple Agents. Why can t we all just get along? (Rodney King) CS 3793/5233 Artificial Intelligence Multiple Agents 1 Multiple Agents Why can t we all just get along? (Rodney King) CS 3793/5233 Artificial Intelligence Multiple Agents 1 Assumptions Assumptions Definitions Partially bservable Each agent can act autonomously.

More information

Constraint Propagation: The Heart of Constraint Programming

Constraint Propagation: The Heart of Constraint Programming Constraint Propagation: The Heart of Constraint Programming Zeynep KIZILTAN Department of Computer Science University of Bologna Email: zeynep@cs.unibo.it URL: http://zeynep.web.cs.unibo.it/ What is it

More information

Topics in Artificial Intelligence: Multiagent Systems Selfish Routing in Computer Networks

Topics in Artificial Intelligence: Multiagent Systems Selfish Routing in Computer Networks Topics in Artificial Intelligence: Multiagent Systems Selfish Routing in Computer Networks Sebastian Streg December 10, 2005 1 Introduction Efficiency in networks with lots of traffic is a serious problem

More information

Simple and Scalable Time-Table Filtering for the Cumulative Constraint

Simple and Scalable Time-Table Filtering for the Cumulative Constraint Simple and Scalable Time-Table Filtering for the Cumulative Constraint Steven Gay, Renaud Hartert, Pierre Schaus UCLouvain, ICTEAM, Place Sainte Barbe 2, 1348 Louvain-la-Neuve, Belgium {firstname.lastname}@uclouvain.be

More information

Strategic Network Formation

Strategic Network Formation Strategic Network Formation Zhongjing Yu, Big Data Research Center, UESTC Email:junmshao@uestc.edu.cn http://staff.uestc.edu.cn/shaojunming What s meaning of Strategic Network Formation? Node : a individual.

More information

Constraint Satisfaction Algorithms for Graphical Games

Constraint Satisfaction Algorithms for Graphical Games Constraint Satisfaction Algorithms for Graphical Games Vishal Soni soniv@umich.edu Satinder Singh baveja@umich.edu Computer Science and Engineering Division University of Michigan, Ann Arbor Michael P.

More information

Game Theory & Networks

Game Theory & Networks Game Theory & Networks (an incredibly brief overview) ndrew Smith ECS 253/ME 289 May 10th, 2016 Game theory can help us answer important questions for scenarios where: players/agents (nodes) are autonomous

More information

A CSP Search Algorithm with Reduced Branching Factor

A CSP Search Algorithm with Reduced Branching Factor A CSP Search Algorithm with Reduced Branching Factor Igor Razgon and Amnon Meisels Department of Computer Science, Ben-Gurion University of the Negev, Beer-Sheva, 84-105, Israel {irazgon,am}@cs.bgu.ac.il

More information

A Fast Arc Consistency Algorithm for n-ary Constraints

A Fast Arc Consistency Algorithm for n-ary Constraints A Fast Arc Consistency Algorithm for n-ary Constraints Olivier Lhomme 1 and Jean-Charles Régin 2 1 ILOG, 1681, route des Dolines, 06560 Valbonne, FRANCE 2 Computing and Information Science, Cornell University,

More information

CIS 192: Artificial Intelligence. Search and Constraint Satisfaction Alex Frias Nov. 30 th

CIS 192: Artificial Intelligence. Search and Constraint Satisfaction Alex Frias Nov. 30 th CIS 192: Artificial Intelligence Search and Constraint Satisfaction Alex Frias Nov. 30 th What is AI? Designing computer programs to complete tasks that are thought to require intelligence 4 categories

More information

PART II: Local Consistency & Constraint Propagation

PART II: Local Consistency & Constraint Propagation PART II: Local Consistency & Constraint Propagation Solving CSPs Search algorithm. Usually backtracking search performing a depth-first traversal of a search tree. Local consistency and constraint propagation.

More information

A New Algorithm for Singleton Arc Consistency

A New Algorithm for Singleton Arc Consistency A New Algorithm for Singleton Arc Consistency Roman Barták, Radek Erben Charles University, Institute for Theoretical Computer Science Malostranské nám. 2/25, 118 Praha 1, Czech Republic bartak@kti.mff.cuni.cz,

More information

Nash Equilibrium Load Balancing

Nash Equilibrium Load Balancing Nash Equilibrium Load Balancing Computer Science Department Collaborators: A. Kothari, C. Toth, Y. Zhou Load Balancing A set of m servers or machines. A set of n clients or jobs. Each job can be run only

More information

Nash equilibria in Voronoi Games on Graphs

Nash equilibria in Voronoi Games on Graphs Nash equilibria in Voronoi Games on Graphs Christoph Dürr, Nguyễn Kim Thắng (Ecole Polytechnique) ESA, Eilat October 07 Plan Motivation : Study the interaction between selfish agents on Internet k players,

More information

Introduction to Algorithms / Algorithms I Lecturer: Michael Dinitz Topic: Algorithms and Game Theory Date: 12/3/15

Introduction to Algorithms / Algorithms I Lecturer: Michael Dinitz Topic: Algorithms and Game Theory Date: 12/3/15 600.363 Introduction to Algorithms / 600.463 Algorithms I Lecturer: Michael Dinitz Topic: Algorithms and Game Theory Date: 12/3/15 25.1 Introduction Today we re going to spend some time discussing game

More information

CSP- and SAT-based Inference Techniques Applied to Gnomine

CSP- and SAT-based Inference Techniques Applied to Gnomine CSP- and SAT-based Inference Techniques Applied to Gnomine Bachelor Thesis Faculty of Science, University of Basel Department of Computer Science Artificial Intelligence ai.cs.unibas.ch Examiner: Prof.

More information

On the Computational Complexity of Nash Equilibria for (0, 1) Bimatrix Games

On the Computational Complexity of Nash Equilibria for (0, 1) Bimatrix Games On the Computational Complexity of Nash Equilibria for (0, 1) Bimatrix Games Bruno Codenotti Daniel Štefankovič Abstract The computational complexity of finding a Nash equilibrium in a nonzero sum bimatrix

More information

Games in Networks: the price of anarchy, stability and learning. Éva Tardos Cornell University

Games in Networks: the price of anarchy, stability and learning. Éva Tardos Cornell University Games in Networks: the price of anarchy, stability and learning Éva Tardos Cornell University Why care about Games? Users with a multitude of diverse economic interests sharing a Network (Internet) browsers

More information

Constraint (Logic) Programming

Constraint (Logic) Programming Constraint (Logic) Programming Roman Barták Faculty of Mathematics and Physics, Charles University in Prague, Czech Republic bartak@ktiml.mff.cuni.cz Sudoku Combinatorial puzzle, whose goal is to enter

More information

CS364A: Algorithmic Game Theory Lecture #19: Pure Nash Equilibria and PLS-Completeness

CS364A: Algorithmic Game Theory Lecture #19: Pure Nash Equilibria and PLS-Completeness CS364A: Algorithmic Game Theory Lecture #19: Pure Nash Equilibria and PLS-Completeness Tim Roughgarden December 2, 2013 1 The Big Picture We now have an impressive list of tractability results polynomial-time

More information

Computing Pure Nash Equilibria in Symmetric Action Graph Games

Computing Pure Nash Equilibria in Symmetric Action Graph Games Computing Pure Nash Equilibria in Symmetric Action Graph Games Albert Xin Jiang Kevin Leyton-Brown Department of Computer Science University of British Columbia {jiang;kevinlb}@cs.ubc.ca Abstract We analyze

More information

CS-171, Intro to A.I. Mid-term Exam Winter Quarter, 2016

CS-171, Intro to A.I. Mid-term Exam Winter Quarter, 2016 CS-171, Intro to A.I. Mid-term Exam Winter Quarter, 016 YOUR NAME: YOUR ID: ID TO RIGHT: ROW: SEAT: The exam will begin on the next page. Please, do not turn the page until told. When you are told to begin

More information

A New Approach to Meusnier s Theorem in Game Theory

A New Approach to Meusnier s Theorem in Game Theory Applied Mathematical Sciences, Vol. 11, 2017, no. 64, 3163-3170 HIKARI Ltd, www.m-hikari.com https://doi.org/10.12988/ams.2017.712352 A New Approach to Meusnier s Theorem in Game Theory Senay Baydas Yuzuncu

More information

Complexity. Congestion Games. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Congestion Games

Complexity. Congestion Games. Algorithmic Game Theory. Alexander Skopalik Algorithmic Game Theory 2013 Congestion Games Algorithmic Game Theory Complexity of pure Nash equilibria We investigate the complexity of finding Nash equilibria in different kinds of congestion games. Our study is restricted to congestion games with

More information

Using HMM in Strategic Games

Using HMM in Strategic Games Using HMM in Strategic Games Mario Benevides Isaque Lima Rafael Nader Pedro Rougemont Systems and Computer Engineering Program and Computer Science Department Federal University of Rio de Janeiro, Brazil

More information

Network Topology and Equilibrium Existence in Weighted Network Congestion Games

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

More information

Minion: Fast, Scalable Constraint Solving. Ian Gent, Chris Jefferson, Ian Miguel

Minion: Fast, Scalable Constraint Solving. Ian Gent, Chris Jefferson, Ian Miguel Minion: Fast, Scalable Constraint Solving Ian Gent, Chris Jefferson, Ian Miguel 1 60 Second Introduction to CSPs Standard Definition A CSP is a tuple V: list of variables D: a domain for each

More information

Algorithms for Computing Strategies in Two-Player Simultaneous Move Games

Algorithms for Computing Strategies in Two-Player Simultaneous Move Games Algorithms for Computing Strategies in Two-Player Simultaneous Move Games Branislav Bošanský 1a, Viliam Lisý a, Marc Lanctot 2b, Jiří Čermáka, Mark H.M. Winands b a Agent Technology Center, Department

More information

Computing Nash Equilibria of Action-Graph Games via Support Enumeration

Computing Nash Equilibria of Action-Graph Games via Support Enumeration Computing Nash Equilibria of Action-Graph Games via Support Enumeration David R. M. Thompson 1, Samantha Leung 2, and Kevin Leyton-Brown 1 1 {daveth, kevinlb}@cs.ubc.ca, University of British Columbia

More information

15-451/651: Design & Analysis of Algorithms October 11, 2018 Lecture #13: Linear Programming I last changed: October 9, 2018

15-451/651: Design & Analysis of Algorithms October 11, 2018 Lecture #13: Linear Programming I last changed: October 9, 2018 15-451/651: Design & Analysis of Algorithms October 11, 2018 Lecture #13: Linear Programming I last changed: October 9, 2018 In this lecture, we describe a very general problem called linear programming

More information

Algorithmic Game Theory and Applications. Lecture 16: Selfish Network Routing, Congestion Games, and the Price of Anarchy.

Algorithmic Game Theory and Applications. Lecture 16: Selfish Network Routing, Congestion Games, and the Price of Anarchy. Algorithmic Game Theory and Applications Lecture 16: Selfish Network Routing, Congestion Games, and the Price of Anarchy Kousha Etessami games and the internet Basic idea: The internet is a huge experiment

More information

An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP

An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP An Extension of Complexity Bounds and Dynamic Heuristics for Tree-Decompositions of CSP Philippe Jégou, Samba Ndojh Ndiaye, and Cyril Terrioux LSIS - UMR CNRS 6168 Université Paul Cézanne (Aix-Marseille

More information

How Bad is Selfish Routing?

How Bad is Selfish Routing? How Bad is Selfish Routing? Tim Roughgarden and Éva Tardos Presented by Brighten Godfrey 1 Game Theory Two or more players For each player, a set of strategies For each combination of played strategies,

More information

Algorithmic Game Theory - Introduction, Complexity, Nash

Algorithmic Game Theory - Introduction, Complexity, Nash Algorithmic Game Theory - Introduction, Complexity, Nash Branislav Bošanský Czech Technical University in Prague branislav.bosansky@agents.fel.cvut.cz February 25, 2018 About This Course main topics of

More information

The Price of Selfishness in Network Coding Jason R. Marden, Member, IEEE, and Michelle Effros, Fellow, IEEE

The Price of Selfishness in Network Coding Jason R. Marden, Member, IEEE, and Michelle Effros, Fellow, IEEE IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 58, NO. 4, APRIL 2012 2349 The Price of Selfishness in Network Coding Jason R. Marden, Member, IEEE, and Michelle Effros, Fellow, IEEE Abstract A game-theoretic

More information

Optimal and Game-Theoretic Deployment of Security Investments in Interdependent Assets

Optimal and Game-Theoretic Deployment of Security Investments in Interdependent Assets Optimal and Game-Theoretic Deployment of Security Investments in Interdependent Assets Ashish R. Hota, Abraham A. Clements, Shreyas Sundaram and Saurabh Bagchi School of Electrical and Computer Engineering

More information

Solving Multi-Objective Distributed Constraint Optimization Problems. CLEMENT Maxime. Doctor of Philosophy

Solving Multi-Objective Distributed Constraint Optimization Problems. CLEMENT Maxime. Doctor of Philosophy Solving Multi-Objective Distributed Constraint Optimization Problems CLEMENT Maxime Doctor of Philosophy Department of Informatics School of Multidisciplinary Sciences SOKENDAI (The Graduate University

More information

Lecture Notes on Congestion Games

Lecture Notes on Congestion Games Lecture Notes on Department of Computer Science RWTH Aachen SS 2005 Definition and Classification Description of n agents share a set of resources E strategy space of player i is S i 2 E latency for resource

More information

MIDTERM EXAMINATION Networked Life (NETS 112) November 21, 2013 Prof. Michael Kearns

MIDTERM EXAMINATION Networked Life (NETS 112) November 21, 2013 Prof. Michael Kearns MIDTERM EXAMINATION Networked Life (NETS 112) November 21, 2013 Prof. Michael Kearns This is a closed-book exam. You should have no material on your desk other than the exam itself and a pencil or pen.

More information

The All Different and Global Cardinality Constraints on Set, Multiset and Tuple Variables

The All Different and Global Cardinality Constraints on Set, Multiset and Tuple Variables The All Different and Global Cardinality Constraints on Set, Multiset and Tuple Variables Claude-Guy Quimper 1 and Toby Walsh 2 1 School of Computer Science, University of Waterloo, Canada, cquimper@math.uwaterloo.ca

More information

An Agent-based Model for the Evolution of the Internet Ecosystem

An Agent-based Model for the Evolution of the Internet Ecosystem An Agent-based Model for the Evolution of the Internet Ecosystem Amogh Dhamdhere Constantine Dovrolis Georgia Tech The Internet Ecosystem 27,000 autonomous networks independently operated and managed The

More information

CONVENTIONAL EXECUTABLE SEMANTICS. Grigore Rosu CS522 Programming Language Semantics

CONVENTIONAL EXECUTABLE SEMANTICS. Grigore Rosu CS522 Programming Language Semantics CONVENTIONAL EXECUTABLE SEMANTICS Grigore Rosu CS522 Programming Language Semantics Conventional Semantic Approaches A language designer should understand the existing design approaches, techniques and

More information

View-based Propagator Derivation

View-based Propagator Derivation View-based Propagator Derivation Christian Schulte SCALE, KTH & SICS, Sweden joint work with: Guido Tack NICTA & Monash University, Australia Based on:. Christian Schulte, Guido Tack. Constraints 18(1),

More information

CONVENTIONAL EXECUTABLE SEMANTICS. Grigore Rosu CS422 Programming Language Semantics

CONVENTIONAL EXECUTABLE SEMANTICS. Grigore Rosu CS422 Programming Language Semantics CONVENTIONAL EXECUTABLE SEMANTICS Grigore Rosu CS422 Programming Language Semantics Conventional Semantic Approaches A language designer should understand the existing design approaches, techniques and

More information

The Price of Selfishness in Network Coding

The Price of Selfishness in Network Coding The Price of Selfishness in Network Coding Jason R. Marden and Michelle Effros Abstract We introduce a game theoretic framework for studying a restricted form of network coding in a general wireless network.

More information

Constraint Satisfaction Problems. Chapter 6

Constraint Satisfaction Problems. Chapter 6 Constraint Satisfaction Problems Chapter 6 Constraint Satisfaction Problems A constraint satisfaction problem consists of three components, X, D, and C: X is a set of variables, {X 1,..., X n }. D is a

More information

Constraint Satisfaction Problems

Constraint Satisfaction Problems Constraint Satisfaction Problems Search and Lookahead Bernhard Nebel, Julien Hué, and Stefan Wölfl Albert-Ludwigs-Universität Freiburg June 4/6, 2012 Nebel, Hué and Wölfl (Universität Freiburg) Constraint

More information

1 Tree Search (12 points)

1 Tree Search (12 points) 1 Tree Search (12 points) Consider the tree shown below. The numbers on the arcs are the arc lengths. Assume that the nodes are expanded in alphabetical order when no other order is specified by the search,

More information

Pathways to Equilibria, Pretty Pictures and Diagrams (PPAD)

Pathways to Equilibria, Pretty Pictures and Diagrams (PPAD) 1 Pathways to Equilibria, Pretty Pictures and Diagrams (PPAD) Bernhard von Stengel partly joint work with: Marta Casetti, Julian Merschen, Lászlo Végh Department of Mathematics London School of Economics

More information

Introduction to Fall 2014 Artificial Intelligence Midterm Solutions

Introduction to Fall 2014 Artificial Intelligence Midterm Solutions CS Introduction to Fall Artificial Intelligence Midterm Solutions INSTRUCTIONS You have minutes. The exam is closed book, closed notes except a one-page crib sheet. Please use non-programmable calculators

More information

2. Modeling AEA 2018/2019. Based on Algorithm Engineering: Bridging the Gap Between Algorithm Theory and Practice - ch. 2

2. Modeling AEA 2018/2019. Based on Algorithm Engineering: Bridging the Gap Between Algorithm Theory and Practice - ch. 2 2. Modeling AEA 2018/2019 Based on Algorithm Engineering: Bridging the Gap Between Algorithm Theory and Practice - ch. 2 Content Introduction Modeling phases Modeling Frameworks Graph Based Models Mixed

More information

Mathematical Programming Formulations, Constraint Programming

Mathematical Programming Formulations, Constraint Programming Outline DM87 SCHEDULING, TIMETABLING AND ROUTING Lecture 3 Mathematical Programming Formulations, Constraint Programming 1. Special Purpose Algorithms 2. Constraint Programming Marco Chiarandini DM87 Scheduling,

More information

Algorithmic Game Theory and Applications. Lecture 16: Selfish Network Routing, Congestion Games, and the Price of Anarchy

Algorithmic Game Theory and Applications. Lecture 16: Selfish Network Routing, Congestion Games, and the Price of Anarchy Algorithmic Game Theory and Applications Lecture 16: Selfish Network Routing, Congestion Games, and the Price of Anarchy Kousha Etessami warning, again 1 In the few remaining lectures, we will briefly

More information

Why Should You Care? Writing a CSP Solver in 3 (or 4) Easy Lessons. Talk Aims. Why Should You Care? Not Talk Aims. Talk Aims

Why Should You Care? Writing a CSP Solver in 3 (or 4) Easy Lessons. Talk Aims. Why Should You Care? Not Talk Aims. Talk Aims Writing a CSP Solver in 3 (or 4) Easy Lessons Christopher Jefferson University of Oxford Why Should ou Care? Understanding how the solver works makes better models. Minion already imposes this on you,

More information

General Methods and Search Algorithms

General Methods and Search Algorithms DM811 HEURISTICS AND LOCAL SEARCH ALGORITHMS FOR COMBINATORIAL OPTIMZATION Lecture 3 General Methods and Search Algorithms Marco Chiarandini 2 Methods and Algorithms A Method is a general framework for

More information

(Due to rounding, values below may be only approximate estimates.) We will supply these numbers as they become available.

(Due to rounding, values below may be only approximate estimates.) We will supply these numbers as they become available. Below, for each problem on this Midterm Exam, Perfect is the percentage of students who received full credit, Partial is the percentage who received partial credit, and Zero is the percentage who received

More information

AUTOMATED REASONING. Agostino Dovier. Udine, October 1, Università di Udine CLPLAB

AUTOMATED REASONING. Agostino Dovier. Udine, October 1, Università di Udine CLPLAB AUTOMATED REASONING Agostino Dovier Università di Udine CLPLAB Udine, October 1, 2018 AGOSTINO DOVIER (CLPLAB) AUTOMATED REASONING UDINE, OCTOBER 1, 2018 1 / 28 COURSE PLACEMENT International Master Degree

More information

Network games. Brighten Godfrey CS 538 September slides by Brighten Godfrey unless otherwise noted

Network games. Brighten Godfrey CS 538 September slides by Brighten Godfrey unless otherwise noted Network games Brighten Godfrey CS 538 September 27 2011 slides 2010-2011 by Brighten Godfrey unless otherwise noted Demo Game theory basics Games & networks: a natural fit Game theory Studies strategic

More information

A Continuation Method for Nash Equilibria in Structured Games

A Continuation Method for Nash Equilibria in Structured Games A Continuation Method for Nash Equilibria in Structured Games Ben Blum Stanford University bblum@stanfordedu Christian R Shelton Stanford University cshelton@csstanfordedu Daphne Koller Stanford University

More information

A Hybrid Constraint Programming Approach to Nurse Rostering Problem

A Hybrid Constraint Programming Approach to Nurse Rostering Problem A Hybrid Constraint Programming Approach to Nurse Rostering Problem Fang He*, Dr. Rong Qu The Automated Scheduling, Optimisation and Planning (ASAP) research group School of Computer Science University

More information

Local Consistency in Weighted CSPs and Inference in Max-SAT

Local Consistency in Weighted CSPs and Inference in Max-SAT Local Consistency in Weighted CSPs and Inference in Max-SAT Student name: Federico Heras Supervisor name: Javier Larrosa Universitat Politecnica de Catalunya, Barcelona, Spain fheras@lsi.upc.edu,larrosa@lsi.upc.edu

More information

Constraint Satisfaction Problems

Constraint Satisfaction Problems Constraint Satisfaction Problems Constraint Optimization Bernhard Nebel, Julien Hué, and Stefan Wölfl Albert-Ludwigs-Universität Freiburg July 17, 2012 Nebel, Hué and Wölfl (Universität Freiburg) Constraint

More information

Introduction to algorithmic mechanism design

Introduction to algorithmic mechanism design Introduction to algorithmic mechanism design Elias Koutsoupias Department of Computer Science University of Oxford EWSCS 2014 March 5-7, 2014 Part I Game Theory and Computer Science Why Game Theory and

More information

A Game Theoretic Approach to Network Coding

A Game Theoretic Approach to Network Coding IW 2009, Volos, Greece, June 10-12, 2009 A Game heoretic Approach to Network Coding Jason R. Marden and Michelle Effros Abstract We introduce a game theoretic framework for studying a restricted form of

More information

The Branch & Move algorithm: Improving Global Constraints Support by Local Search

The Branch & Move algorithm: Improving Global Constraints Support by Local Search Branch and Move 1 The Branch & Move algorithm: Improving Global Constraints Support by Local Search Thierry Benoist Bouygues e-lab, 1 av. Eugène Freyssinet, 78061 St Quentin en Yvelines Cedex, France tbenoist@bouygues.com

More information

Constraint Satisfaction Problems

Constraint Satisfaction Problems Constraint Satisfaction Problems In which we see how treating states as more than just little black boxes leads to the invention of a range of powerful new search methods and a deeper understanding of

More information

THE CONSTRAINT PROGRAMMER S TOOLBOX. Christian Schulte, SCALE, KTH & SICS

THE CONSTRAINT PROGRAMMER S TOOLBOX. Christian Schulte, SCALE, KTH & SICS THE CONSTRAINT PROGRAMMER S TOOLBOX Christian Schulte, SCALE, KTH & SICS Constraint Programming 2 What is constraint programming? Sudoku is constraint programming 3 Sudoku...is constraint programming!

More information

Search. Krzysztof Kuchcinski. Department of Computer Science Lund Institute of Technology Sweden.

Search. Krzysztof Kuchcinski. Department of Computer Science Lund Institute of Technology Sweden. Search Krzysztof Kuchcinski Krzysztof.Kuchcinski@cs.lth.se Department of Computer Science Lund Institute of Technology Sweden January 12, 2015 Kris Kuchcinski (LTH) Search January 12, 2015 1 / 46 Search

More information

Pascal De Beck-Courcelle. Master in Applied Science. Electrical and Computer Engineering

Pascal De Beck-Courcelle. Master in Applied Science. Electrical and Computer Engineering Study of Multiple Multiagent Reinforcement Learning Algorithms in Grid Games by Pascal De Beck-Courcelle A thesis submitted to the Faculty of Graduate and Postdoctoral Affairs in partial fulfillment of

More information

On the implementation of a multiple output algorithm for defeasible argumentation

On the implementation of a multiple output algorithm for defeasible argumentation On the implementation of a multiple output algorithm for defeasible argumentation Teresa Alsinet 1, Ramón Béjar 1, Lluis Godo 2, and Francesc Guitart 1 1 Department of Computer Science University of Lleida

More information

Constraint Programming

Constraint Programming Depth-first search Let us go back to foundations: DFS = Depth First Search Constraint Programming Roman Barták Department of Theoretical Computer Science and Mathematical Logic 2 3 4 5 6 7 8 9 Observation:

More information

Branch & Bound (B&B) and Constraint Satisfaction Problems (CSPs)

Branch & Bound (B&B) and Constraint Satisfaction Problems (CSPs) Branch & Bound (B&B) and Constraint Satisfaction Problems (CSPs) Alan Mackworth UBC CS 322 CSP 1 January 25, 2013 P&M textbook 3.7.4 & 4.0-4.2 Lecture Overview Recap Branch & Bound Wrap up of search module

More information

Chapter 3: Search. c D. Poole, A. Mackworth 2010, W. Menzel 2015 Artificial Intelligence, Chapter 3, Page 1

Chapter 3: Search. c D. Poole, A. Mackworth 2010, W. Menzel 2015 Artificial Intelligence, Chapter 3, Page 1 Chapter 3: Search c D. Poole, A. Mackworth 2010, W. Menzel 2015 Artificial Intelligence, Chapter 3, Page 1 Searching Often we are not given an algorithm to solve a problem, but only a specification of

More information

A Game-Theoretic Framework for Congestion Control in General Topology Networks

A Game-Theoretic Framework for Congestion Control in General Topology Networks A Game-Theoretic Framework for Congestion Control in General Topology SYS793 Presentation! By:! Computer Science Department! University of Virginia 1 Outline 2 1 Problem and Motivation! Congestion Control

More information

Constraint Satisfaction Problems

Constraint Satisfaction Problems Constraint Satisfaction Problems Tuomas Sandholm Carnegie Mellon University Computer Science Department [Read Chapter 6 of Russell & Norvig] Constraint satisfaction problems (CSPs) Standard search problem:

More information

10703 Deep Reinforcement Learning and Control

10703 Deep Reinforcement Learning and Control 10703 Deep Reinforcement Learning and Control Russ Salakhutdinov Machine Learning Department rsalakhu@cs.cmu.edu Policy Gradient I Used Materials Disclaimer: Much of the material and slides for this lecture

More information

Stochastic Coalitional Games with Constant Matrix of Transition Probabilities

Stochastic Coalitional Games with Constant Matrix of Transition Probabilities Applied Mathematical Sciences, Vol. 8, 2014, no. 170, 8459-8465 HIKARI Ltd, www.m-hikari.com http://dx.doi.org/10.12988/ams.2014.410891 Stochastic Coalitional Games with Constant Matrix of Transition Probabilities

More information

Repairing Multi-Player Games

Repairing Multi-Player Games Repairing Multi-Player Games Shaull Almagor, Guy Avni, and Orna Kupferman School of Computer Science and Engineering, The Hebrew University, Israel Abstract Synthesis is the automated construction of systems

More information

Incremental Policy Generation for Finite-Horizon DEC-POMDPs

Incremental Policy Generation for Finite-Horizon DEC-POMDPs Incremental Policy Generation for Finite-Horizon DEC-POMDPs Chistopher Amato Department of Computer Science University of Massachusetts Amherst, MA 01003 USA camato@cs.umass.edu Jilles Steeve Dibangoye

More information

6.890: Fun with Hardness Proofs

6.890: Fun with Hardness Proofs 6.890: Fun with Hardness Proofs Guest Lectures on PPAD-Part B November 2014 Constantinos Daskalakis CSAIL, MIT Last time Existence Theorems: Nash, Brouwer, Sperner Total Search Problems in NP Totality

More information

Crossword Puzzles as a Constraint Problem

Crossword Puzzles as a Constraint Problem Crossword Puzzles as a Constraint Problem Anbulagan and Adi Botea NICTA and Australian National University, Canberra, Australia {anbulagan,adi.botea}@nicta.com.au Abstract. We present new results in crossword

More information

Representation of Finite Games as Network Congestion Games

Representation of Finite Games as Network Congestion Games Representation of Finite Games as Network Congestion Games Igal Milchtaich To cite this version: Igal Milchtaich. Representation of Finite Games as Network Congestion Games. Roberto Cominetti and Sylvain

More information

Artificial Intelligence (part 4a) Problem Solving Using Search: Structures and Strategies for State Space Search

Artificial Intelligence (part 4a) Problem Solving Using Search: Structures and Strategies for State Space Search Artificial Intelligence (part 4a) Problem Solving Using Search: Structures and Strategies for State Space Search Course Contents Again..Selected topics for our course. Covering all of AI is impossible!

More information

Path-planning for Multiple Robots

Path-planning for Multiple Robots Path-planning for Multiple Robots RNDr. Pavel Surynek, Ph.D. Department of Theoretical Computer Science and Mathematical Logic Faculty of Mathematics and Physics Charles University in Prague http://ktiml.mff.cuni.cz/~surynek

More information

Combining finite and continuous solvers Towards a simpler solver maintenance

Combining finite and continuous solvers Towards a simpler solver maintenance Combining finite and continuous solvers Towards a simpler solver maintenance Jean-Guillaume Fages, Gilles Chabert, Charles Prud Homme To cite this version: Jean-Guillaume Fages, Gilles Chabert, Charles

More information

Search: Advanced Topics and Conclusion

Search: Advanced Topics and Conclusion Search: Advanced Topics and Conclusion CPSC 322 Lecture 8 January 24, 2007 Textbook 2.6 Search: Advanced Topics and Conclusion CPSC 322 Lecture 8, Slide 1 Lecture Overview 1 Recap 2 Branch & Bound 3 A

More information

Constraint Satisfaction Problems

Constraint Satisfaction Problems Constraint Satisfaction Problems CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2013 Soleymani Course material: Artificial Intelligence: A Modern Approach, 3 rd Edition,

More information

On a Network Generalization of the Minmax Theorem

On a Network Generalization of the Minmax Theorem On a Network Generalization of the Minmax Theorem Constantinos Daskalakis Christos H. Papadimitriou {costis, christos}@cs.berkeley.edu February 10, 2009 Abstract We consider graphical games in which edges

More information

Graphical Models for Game Theory

Graphical Models for Game Theory Graphical Models for Game Theory Michael Kearns Michael L. Littman Satinder Singh AT&T Labs Florham Park, New Jersey fmkearns,mlittman,bavejag@research.att.com Abstract We introduce a compact graph-theoretic

More information

Constraint Programming

Constraint Programming Constraint Programming - An overview Examples, Satisfaction vs. Optimization Different Domains Constraint Propagation» Kinds of Consistencies Global Constraints Heuristics Symmetries 7 November 0 Advanced

More information

CS-171, Intro to A.I. Mid-term Exam Fall Quarter, 2014

CS-171, Intro to A.I. Mid-term Exam Fall Quarter, 2014 CS-171, Intro to A.I. Mid-term Exam Fall Quarter, 2014 YOUR NAME: YOUR ID: ID TO RIGHT: ROW: SEAT: The exam will begin on the next page. Please, do not turn the page until told. When you are told to begin

More information

Heuristic Search and Advanced Methods

Heuristic Search and Advanced Methods Heuristic Search and Advanced Methods Computer Science cpsc322, Lecture 3 (Textbook Chpt 3.6 3.7) May, 15, 2012 CPSC 322, Lecture 3 Slide 1 Course Announcements Posted on WebCT Assignment1 (due on Thurs!)

More information