game dynamics in extensive form
play

Game Dynamics in Extensive Form Dietmar Berwanger LSV, ENS Cachan - PowerPoint PPT Presentation

Sep 18, 2022 •165 likes •634 views

Game Dynamics in Extensive Form Dietmar Berwanger LSV, ENS Cachan & CNRS ICLA - Logic & Social Interaction, 2009 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 1 / 13 Evolution in repeated games 0,1 2,1 1,2

  1. Game Dynamics in Extensive Form Dietmar Berwanger LSV, ENS Cachan & CNRS ICLA - Logic & Social Interaction, 2009 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 1 / 13

  2. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  3. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  4. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  5. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  6. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  7. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  8. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  9. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  10. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  11. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  12. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  13. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  14. Evolution in repeated games 0,1 2,1 1,2 0,3 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  15. Evolution in repeated games Justification of Nash Equilibrium: 0,1 2,1 definite state 1,2 0,3 selective pressure 1,1 2,0 ◮ sequential dynamics Project: Reveal extensive structure from the normal-form representation of a game. Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 2 / 13

  16. Game dynamics graph 0,1 2,1 0,1 2,1 Game Γ in normal form: 1,2 0,3 1,2 0,3 n players 1,1 2,0 1,1 2,0 strategy sets S i -- finite utility u i : S → N 0,1 2,1 0,1 2,1 1,2 0,3 1,2 0,3 ◮ Game dynamics graph G ( Γ ) : 1,1 2,0 1,1 2,0 nodes: profiles in S edges: switches s → s ′ 0,1 2,1 0,1 2,1 by any subset of players 1,2 0,3 1,2 0,3 1,1 2,0 1,1 2,0 We consider pure strategies. Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 3 / 13

  17. Best-response dynamics 0,1 2,1 0,1 2,1 1,2 0,3 1,2 0,3 1,1 2,0 1,1 2,0 Greedy walks converge to Nash Equilibrium 0,1 2,1 0,1 2,1 if well-founded 1,2 0,3 1,2 0,3 e.g., in potential games. 1,1 2,0 1,1 2,0 0,1 2,1 0,1 2,1 1,2 0,3 1,2 0,3 1,1 2,0 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 4 / 13

  18. Best-response dynamics 0,1 2,1 0,1 2,1 1,2 0,3 1,2 0,3 1,1 2,0 1,1 2,0 Greedy walks converge to Nash Equilibrium 0,1 2,1 0,1 2,1 if well-founded 1,2 0,3 1,2 0,3 e.g., in potential games. 1,1 2,0 1,1 2,0 What if not? 0,1 2,1 0,1 2,1 1,2 0,3 1,2 0,3 1,1 2,0 1,1 2,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 4 / 13

  19. Sink equilibria [Vetta & al, 2005] [Goemans, Mirokni, Vetta 2005] Sink: terminal connected component of best-response graph. ◮ Price of sinking -- social cost of lack of coordination vs price of anarchy. Σu i ( ♦♣t ) Σu i ( ♦♣t ) Σu i ( ✇♦rst ❙✐♥❦❊q ) � Σu i ( ✇♦rst ◆❛s❤❊q ) Theorem. The price of anarchy can underestimate the price of sinking by a factor of n . Consequences for convergence speed of random best-response walks. Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 5 / 13

  20. Salience of the definite state Example: Tragedy of the commons - variant each player has one responsible and n irresponsible strategies ◮ responsible strategies guarantee 1 Rp ◮ irresponsible strategies pay off 2 Rs for one player ◮ all other irresponsibles -- 0 Rs who wins depends on all chosen strategies. Pareto optimum: n+1 · Nash Eq: n+ ǫ · Sink Eq: 2 ◮ Effect hides when mixing strategies relies on perfect information about the current state Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 6 / 13

  21. Strategising equilibrium selection 1,1 0,0 1,1 0,0 0,0 2,1 0,0 2,1 1,1 0,0 1,1 0,0 0,0 2,1 0,0 2,1 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 7 / 13

  22. Strategising equilibrium selection 1,1 0,0 1,1 0,0 0,0 2,1 0,0 2,1 1,1 0,0 1,1 0,0 0,0 2,1 0,0 2,1 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 7 / 13

  23. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  24. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  25. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  26. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  27. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  28. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  29. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  30. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  31. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  32. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  33. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  34. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  35. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  36. Efficient orbit supported by a non-greedy switch 1,1,1 0,0,0 1,1,0 2,0,1 0,0,0 2,2,2 1,2,1 1,3,3 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 8 / 13

  37. Dynamics metagame Game graph: G ( Γ ) ; 0,0 0,0 utility: cumulative ◮ mean payoff, discounted 0,1 0,0 perfect information ◮ no procedural rules ◮ fair tie breaking 0,0 1,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 9 / 13

  38. Dynamics metagame Game graph: G ( Γ ) ; 0,0 0,0 utility: cumulative ◮ mean payoff, discounted 0,1 0,0 perfect information ◮ no procedural rules ◮ fair tie breaking 0,0 1,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 9 / 13

  39. Basic case: two players 0,0 0,0 0,1 0,1 0,0 0,0 0,0 0,0 0,0 0,0 1,0 1,0 Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 10 / 13

  40. Basic case: two players 0,0 0,0 0,1 0,1 0,0 0,0 0,0 0,0 0,0 0,0 1,0 1,0 Theorem. [Ehrenfeucht, Mycielski] Mean-payoff zero-sum games are determined with memoryless strategies. feasible outcomes ◮ ◭ Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 10 / 13

  41. Basic case: two players 0,0 0,0 0,1 0,1 0,0 0,0 0,0 0,0 0,0 0,0 1,0 1,0 Theorem. [Ehrenfeucht, Mycielski] Mean-payoff zero-sum games are determined with memoryless strategies. feasible outcomes ◮ ◭ Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 10 / 13

  42. Basic case: two players 0,0 0,0 0,1 0,1 0,0 0,0 0,0 0,0 0,0 0,0 1,0 1,0 Theorem. [Ehrenfeucht, Mycielski] Mean-payoff zero-sum games are determined with memoryless strategies. feasible outcomes ◮ ◭ Dietmar Berwanger (CNRS) Game Dynamics Logic & Social Interaction 10 / 13

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Normal Form Games 2-12-16 Game Representations Extensive Form Game Normal Form Game

Normal Form Games 2-12-16 Game Representations Extensive Form Game Normal Form Game

Normal Form Games 2-12-16 Game Representations Extensive Form Game Normal Form Game Agents take turns Agents move simultaneously. Represented as a tree. Represented as a matrix. Internal nodes are agent decisions.

563 views • 12 slides
Extensive Form Games Game Theory MohammadAmin Fazli Algorithmic Game Theory 1 TOC Perfect

Extensive Form Games Game Theory MohammadAmin Fazli Algorithmic Game Theory 1 TOC Perfect

Extensive Form Games Game Theory MohammadAmin Fazli Algorithmic Game Theory 1 TOC Perfect Information Extensive Form Games Backward Induction and MinMax Algorithms Imperfect Information Extensive Form Games The Sequence Form

521 views • 32 slides
Low-Variance and Zero-Variance Baselines in Extensive-Form Games Trevor Davis 2,* , Martin Schmid

Low-Variance and Zero-Variance Baselines in Extensive-Form Games Trevor Davis 2,* , Martin Schmid

1 2 Low-Variance and Zero-Variance Baselines in Extensive-Form Games Trevor Davis 2,* , Martin Schmid 1 , Michael Bowling 1,2 *Work done during internship at DeepMind Monte Carlo game solving Extensive-form games (EFGs) Monte Carlo game

441 views • 21 slides
Game Theory Extensive Form Games Levent Ko ckesen Ko c University Levent Ko ckesen

Game Theory Extensive Form Games Levent Ko ckesen Ko c University Levent Ko ckesen

page.1 Game Theory Extensive Form Games Levent Ko ckesen Ko c University Levent Ko ckesen (Ko c University) Extensive Form Games 1 / 20 page.2 Extensive Form Games Strategic form games are used to model situations in which

948 views • 20 slides
Game Theory Extensive Form Games: Applications Levent Ko ckesen Ko c University Levent

Game Theory Extensive Form Games: Applications Levent Ko ckesen Ko c University Levent

page.1 Game Theory Extensive Form Games: Applications Levent Ko ckesen Ko c University Levent Ko ckesen (Ko c University) Extensive Form Games: Applications 1 / 23 page.2 A Simple Game You have 10 TL to share A makes an offer

384 views • 23 slides
Roadmap 1 Introduction 2 Static games 3 Extensive-form games 4 Summary 2 Introduction to Game

Roadmap 1 Introduction 2 Static games 3 Extensive-form games 4 Summary 2 Introduction to Game

An introduction on game theory for wireless networking [1] Ning Zhang 14 May, 2012 [1] Game Theory in Wireless Networks: A Tutorial 1 Roadmap 1 Introduction 2 Static games 3 Extensive-form games 4 Summary 2 Introduction to Game Theory

321 views • 28 slides
Extensive Form Games Mihai Manea MIT Extensive-Form Games N : finite set of players; nature

Extensive Form Games Mihai Manea MIT Extensive-Form Games N : finite set of players; nature

Extensive Form Games Mihai Manea MIT Extensive-Form Games N : finite set of players; nature is player 0 N tree: order of moves payoffs for every player at the terminal nodes information partition actions available at

656 views • 34 slides
Game Theory Extensive Form Games with Incomplete Information Levent Ko ckesen Ko c

Game Theory Extensive Form Games with Incomplete Information Levent Ko ckesen Ko c

page.1 Game Theory Extensive Form Games with Incomplete Information Levent Ko ckesen Ko c University Levent Ko ckesen (Ko c University) Signaling Games 1 / 27 page.2 Extensive Form Games with Incomplete Information We have seen

517 views • 27 slides
Extensive Form Games Extensive-form games with perfect information When moving, each player

Extensive Form Games Extensive-form games with perfect information When moving, each player

Extensive Form Games Extensive-form games with perfect information When moving, each player is aware of all previous moves 2,10 (perfect information). 3,8 A (pure) strategy for player i is a mapping from player is nodes to 2,10

238 views • 13 slides
Introduction to Game Theory Mehdi Dastani BBL-521 M.M.Dastani@uu.nl Extensive Games

Introduction to Game Theory Mehdi Dastani BBL-521 M.M.Dastani@uu.nl Extensive Games

Introduction to Game Theory Mehdi Dastani BBL-521 M.M.Dastani@uu.nl Extensive Games Sequential Structure of Games. Perfect and Imperfect-Information Extensive Games. Strategies and Equilibria for Extensive Games. Extensive Games

531 views • 28 slides
Advanced Microeconomics: Game Theory P . v. Mouche Wageningen University 2017 Motivation

Advanced Microeconomics: Game Theory P . v. Mouche Wageningen University 2017 Motivation

Motivation Games in strategic form Games in extensive form Advanced Microeconomics: Game Theory P . v. Mouche Wageningen University 2017 Motivation Games in strategic form Games in extensive form Outline Motivation 1 Games in strategic

866 views • 57 slides
Advanced Microeconomics: Game Theory P . v. Mouche Wageningen University 2018 Motivation

Advanced Microeconomics: Game Theory P . v. Mouche Wageningen University 2018 Motivation

Motivation Games in strategic form Games in extensive form Advanced Microeconomics: Game Theory P . v. Mouche Wageningen University 2018 Motivation Games in strategic form Games in extensive form Outline Motivation 1 Games in strategic

683 views • 57 slides
Game Theory P . v. Mouche Wageningen University 2020, Period 4 Organisation Motivation Games

Game Theory P . v. Mouche Wageningen University 2020, Period 4 Organisation Motivation Games

Organisation Motivation Games in strategic form Games in extensive form Game Theory P . v. Mouche Wageningen University 2020, Period 4 Organisation Motivation Games in strategic form Games in extensive form Outline 1 Organisation

747 views • 48 slides
EC487 Advanced Microeconomics, Part I: Lecture 8 Leonardo Felli 32L.LG.04 17 November 2017

EC487 Advanced Microeconomics, Part I: Lecture 8 Leonardo Felli 32L.LG.04 17 November 2017

EC487 Advanced Microeconomics, Part I: Lecture 8 Leonardo Felli 32L.LG.04 17 November 2017 Extensive Form Game Consider the following extensive form game: 1 U D

873 views • 61 slides
ECE700.07: Game Theory with Engineering Applications Le Lecture 5: 5: Ga Games in Ext Extensi

ECE700.07: Game Theory with Engineering Applications Le Lecture 5: 5: Ga Games in Ext Extensi

ECE700.07: Game Theory with Engineering Applications Le Lecture 5: 5: Ga Games in Ext Extensi ensive e Form Seyed Majid Zahedi Outline Perfect information extensive form games Subgame perfect equilibrium Backward induction

687 views • 42 slides
A game in normal form consists of: Set of players = {1, , } Strategy set

A game in normal form consists of: Set of players = {1, , } Strategy set

T RUTH J USTICE A LGOS Game Theory I: Basic Concepts Teachers: Ariel Procaccia (this time) and Alex Psomas NORMAL-FORM GAME A game in normal form consists of: Set of players = {1, , } Strategy set For each

239 views • 21 slides
Game Theory for Sequential Interactions CMPUT 366: Intelligent Systems S&LB 5.0-5.2.2

Game Theory for Sequential Interactions CMPUT 366: Intelligent Systems S&LB 5.0-5.2.2

Game Theory for Sequential Interactions CMPUT 366: Intelligent Systems S&LB 5.0-5.2.2 Lecture Outline 1. Recap 2. Perfect Information Games 3. Backward Induction 4. Imperfect Information Games Recap: Game Theory Game

426 views • 21 slides
IR to RIMS Transforming an Institutional Repository Into a Research Information Management

IR to RIMS Transforming an Institutional Repository Into a Research Information Management

IR to RIMS Transforming an Institutional Repository Into a Research Information Management System Craig Feldman & Darryl Meyer Supervised by Prof Hussein Suleman 1 Introduction (1) Migrate two NRF databases to DSpace Current and

515 views • 24 slides
r s t

r s t

r s t sr P rstt tt

813 views • 16 slides
Substructural modal logic for optimality and games Gabrielle Anderson University College London

Substructural modal logic for optimality and games Gabrielle Anderson University College London

Substructural modal logic for optimality and games Gabrielle Anderson University College London (Joint work with David Pym) Resource Reasoning Wednesday 13th January, 2016 Overview Focus: logical characterisations of notions of

327 views • 19 slides
Single agent or multiple agents Many domains are characterized by multiple agents rather than a

Single agent or multiple agents Many domains are characterized by multiple agents rather than a

Single agent or multiple agents Many domains are characterized by multiple agents rather than a single agent. Game theory studies what agents should do in a multi-agent setting. Agents can be cooperative, competitive or somewhere in between.

883 views • 43 slides
The Role of Commitment in Repeated Games Julio Gonz alez-D az Ignacio Garc a-Jurado

The Role of Commitment in Repeated Games Julio Gonz alez-D az Ignacio Garc a-Jurado

The Role of Commitment in Repeated Games Julio Gonz alez-D az Ignacio Garc a-Jurado Department of Statistics and Operations Research School of Mathematics Universidade de Santiago de Compostela Introduction Outline Motivation

1.47k views • 123 slides
PANEL A: Price of coltan PANEL B: Price of gold Figure 3: Support villages in the sample Notes :

PANEL A: Price of coltan PANEL B: Price of gold Figure 3: Support villages in the sample Notes :

PANEL A: Price of coltan PANEL B: Price of gold Figure 3: Support villages in the sample Notes : This map presents the location of the support villages. Support villages with at least one coltan mine are marked with a circle and cross (red).

419 views • 8 slides
Voting versus Lobbying David K. Levine, Andrea Mattozzi and Salvatore Modica 1 The Setting

Voting versus Lobbying David K. Levine, Andrea Mattozzi and Salvatore Modica 1 The Setting

Voting versus Lobbying David K. Levine, Andrea Mattozzi and Salvatore Modica 1 The Setting political contest between two groups providing or promising effort lobbying groups, political parties consider different mechanisms for

828 views • 26 slides

More recommend