COMPUTATIONALSOCIAL SCIENCE
SEMINAR
HEINRICH H. NAX ([email protected])COSS, ETH ZURICH
SEE WWW.NAX.SCIENCE
TWO PARTSadmin some game
theory intro
of 39
ADMIN
• Information about the course, • Updated materials/slides of speakers, • Program, links, etc
• will be made available at • https://coss.ethz.ch/education/computational.html
• ETH students: please contact me (Heinrich) under [email protected] or Nino under [email protected] with questions about the course!
• Information about the course, • materials/slides of speakers, • Links, etc
• will be made available at • https://coss.ethz.ch/education/computational.html
• ETH students: please contact me (Heinrich) under [email protected] or Nino under [email protected] with questions about the course!
The course is co-organized by
Nino Antulov-Fantulin (Data Science, Machine Learning)https://coss.ethz.ch/people/postdocs/nantulovfantulin.html
We are part of the chair ofDirk Helbing (COSS, ETH) … who is on sabbatical
• Computational Social Science aims at• bringing modeling and computer simulation of social processes and
phenomena together with related empirical, experimental, and data-driven work
• combining perspectives of different scientific disciplines (e.g. socio-physics, social, computer and complexity science)
• bridging between fundamental and applied work using game theory to model individual-level decision-making
+ me
of 39
PRELIMINARY SCHEDULE…
THANKS TO SPEAKERS
“COURSEWORK”Please prepare, in “diverse” groups of 4 – 8 students (less is not encouraged, neither is more),
1. Highly polished presentation in 15-20 slides (which you mightbe asked to hold)
2. Accompanying notes/essay outlining your arguments ofthe presentation including at least one “data”component (max. 60 pages).
Topics: You choose! You may include one or a combination of issues raised during the course, may prove a good understanding of recurring topics covered during the week, combine several ideas, or propose fresh thoughts related to issues raised in the course,… in English or German.
Deadline: December 12!December 1: Propose a project with 1-2 slides, and finalize teams (including one team leader responsible for project delivery).
• A mix of brief introductory and research talks (hopefully exciting)• introducing concepts and methods• some review and research• new ideas
• Each talk will come with some discussion time and we encourage active participation and questioning (unless the speaker says otherwise) – i.e. questions or comments at any moment in time!
• Also use the breaks to form “teams”
YOUR MARKS
WWW.DVSN.APP
SHARING GAINS FROM A JOINT VENTURE
of 39
THE BASIC PROBLEM• n people bake a cake together• the cake is worth 1 dollar
• how should it be split amongst the people?
of 39
1/N?
of 39
1/N?• What if some people did more than others?
of 39
RELATIVE EFFORTS?• Who know this?• Who can verify it?
of 39
CONSIDER THE SCENARIO• n people bake a cake together• the cake is worth 1 dollar• a third party holds it but has no idea of who did what• people submit proposals about how it should be split• the third party aggregates these proposals and pays
of 39
HOW?
of 39
HOW?
of 39
• Please form teams of 4-5 people and think about this for 5 min
• Then every team will have 1 min to present their proposal
“THE SOLUTION”
of 39
“THE SOLUTION”
of 39
3 properties
1. Strategy-proof: requires that your own claim about yourself does not matter for your own share
2. Objective: requires that a partner be unable to affect the share of any other partner by reporting a different belief about his own contribution
3. Consensual: if there is a set of shares that is consistent with all of the inputs that the partners provide, the rule needs to assign those shares.
ILLUSTRATION OF THE FORMULA
of 39
ILLUSTRATION OF THE FORMULA
of 39
ILLUSTRATION OF THE FORMULA
of 39
ILLUSTRATION OF THE FORMULA
of 39
ILLUSTRATION OF THE FORMULA
of 39
ILLUSTRATION OF THE FORMULA
of 39
residual
ILLUSTRATION OF THE FORMULA
of 39
residual
ILLUSTRATION OF THE FORMULA
of 39
residuali
j
k
l
THE FORMULA
of 39
Average relative contribution jk
Average RC jk without i’s opinion
Auxiliary function assigning share to i when j excluded
Final payment
share in the other slicesi’s residual in his slice
THINK ABOUT IT
of 39
AND READ R THEN L THEN R
of 39
EXAMPLE (MADE SIMPLE)
of 39
L 33, 33, 33
ML 50, 25, 25
MR 50, 25, 25
R 50, 25, 25
OUTPUT – NOTE CONSENSUALITY MUST BITE HERE
of 39
say\get L ML MR R
L 33 33 33
ML 50 25 25
MR 50 25 25
R 50 25 25
Gets 40 20 20 20
JUST AVERAGING WOULD GIVE
of 39
say\get L ML MR R
L 33 33 33
ML 50 25 25
MR 50 25 25
R 50 25 25
Gets 50/133 28/133 28/133 28/133=37 =21 =21 =21
DVSN.APP
of 39
THE SITUATION• E.g. 5 students do a course project together• the project gets –for example- a 5.5
• What should the individual marks be?
of 39
5.5 TO ALL?
of 39
5.5?• What if some people did more than others?
of 39
RELATIVE EFFORTS?• Who know this? The examiner doesn’t.• Who can verify it? The examiner cannot.
of 39
YOU DO!• You each specify what the contributions of everyone were
of 39
I USE
of 39
FINAL GRADE
of 39
2 parts
1. The total is group size times group mark
2. Individual marks are based on the mutual evaluation exercise based on this mechanism by de Clippel et al. via www.DVSN.app
GOOD LUCK.
of 39
SOME GAME THEORY
Lecture 1: Introduction
A game
Rules:
1 Players: All of you:
IKUuwQ2 Actions: Choose a number between 0 and 1003 Outcome: The player with the number closest to half the average of all
submitted numbers wins.
4 Payoffs: He/she will receive CHF, which I will pay out
right after the game.
5 In case of several winners, divide payment by number of winners andpay all winners.
2 / 47
Lecture 1: Introduction
Others before you did
3 / 47
Lecture 1: Introduction
Soziologisches Institut
“…It is not a case of choosing those [faces] that, to the best of one’s judgment, are really the prettiest, nor even those that average opinion genuinely thinks the prettiest. We have reached the third degree where we devote our intelligences to anticipating what average opinion expects the average opinion to be. And there are some, I believe, who practice the fourth, fifth and higher degrees.” (John Maynard Keynes, General Theory of Employment, Interest, and Money, 1936, p.156).
Analogy between stock markets and newspaper contest in which people guess what faces others will guess are most beautiful.
Lecture 1: Introduction
Lecture 1: Introduction
Soziologisches Institut
Diekmann, Andreas. "Rational choice, evolution and the “Beauty Contest”." Raymond Boudon. A Life in Sociology. Oxford: Bardwell (2009), p,.8 ff.
K=0K=1K=2K=3
Beliefs and learning
Lecture 1: Introduction
Level 0 (“no reasoning”)random guess or simple rules
Level 1 reacts to base strategy at level 0Guesses of 50 = 33Level 2 reacts to level 1Guesses of of 50 = 22Level k reacts to level k-1
Guesses ( ) k 0…
: Cognitive Hierarchy Theory
Soziologisches Institut
Diekmann, Andreas. "Rational choice, evolution and the “Beauty Contest”." Raymond Boudon. A Life in Sociology. Oxford: Bardwell (2009), p,.8 ff.
K=0K=1K=2K=3
Beliefs and learning
Soziologisches Institut
Diekmann, Andreas. "Rational choice, evolution and the “Beauty Contest”." Raymond Boudon. A Life in Sociology. Oxford: Bardwell (2009), p,.8 ff.
K=0K=1K=2K=3
Beliefs and learning
Soziologisches Institut
Diekmann, Andreas. "Rational choice, evolution and the “Beauty Contest”." Raymond Boudon. A Life in Sociology. Oxford: Bardwell (2009), p,.8 ff.
K=0K=1K=2K=3
Beliefs and learning
Soziologisches Institut
Diekmann, Andreas. "Rational choice, evolution and the “Beauty Contest”." Raymond Boudon. A Life in Sociology. Oxford: Bardwell (2009), p,.8 ff.
K=0K=1K=2K=3
Beliefs and learning
Soziologisches Institut
Bosch-Domènech et al. (2002, AEA)
Soziologisches Institut
Bosch-Domènech et al. (2002, AEA)
Lecture 1: Introduction
Lecture 1: Introduction
Acknowledgments
Bary Pradelski (ETHZ)
Peyton Young (Oxford, LSE)
Bernhard von Stengel (LSE)
Francoise Forges (Paris Dauphine)
Paul Duetting (LSE)
Jeff Shamma (Georgia Tech, KAUST)
Joergen Weibull (Stockholm, TSE)
Andreas Diekmann (ETHZ)
Dirk Helbing (ETHZ)
5 / 47
Lecture 1: Introduction
Game theory
A tour of its people, applications and concepts
1 von Neumann
2 Nash
3 Aumann, Schelling, Selten, Shapley
4 Today
6 / 47
Lecture 1: Introduction
John von Neumann (1903-1957)7 / 47
Lecture 1: Introduction
What is game theory?
A mathematical language to express models of, as Myerson says:
“conflict and cooperation between intelligent rational decision-makers”
In other words, interactive decision theory (Aumann)
Dates back to von Neumann & Morgenstern (1944)
Most important solution concept: the Nash (1950) equilibrium
8 / 47
Lecture 1: Introduction
Games and Non-Games
What is a game? And what is not a game?
9 / 47
Lecture 1: Introduction
Uses of game theory
Prescriptive agenda versus descriptive agenda
“Reverse game theory”/mechanism design:
“in a design problem, the goal function is the main given, while the
mechanism is the unknown.” (Hurwicz)
The mechanism designer is a game designer. He studies
What agents would do in various games
And what game leads to the outcomes that are most desirable
10 / 47
Lecture 1: Introduction
Game theory revolutionized several disciplines
Biology (evolution, conflict, etc.)
Social sciences (economics, sociology, political science, etc.)
Computer science (algorithms, control, etc.)
game theory is now applied widely (e.g. regulation, online auctions,
distributed control, medical research, etc.)
11 / 47
Lecture 1: Introduction
Its impact in economics (evaluated by Nobel prizes)
1972: Ken Arrow − general equilibrium
1994: John Nash, Reinhard Selten, John Harsanyi − solution concepts
2005: Tom Schelling and Robert Aumann − evolutionary game theory
and common knowledge
2007: Leonid Hurwicz, Eric Maskin, Roger Myerson − mechanism
design
2009: Lin Ostrom − economic governance, the commons
2012: Al Roth and Lloyd Shapley − market design
2014: Jean Tirole − markets and regulation
2016: Oliver Hart and Bengt Holmström − contract theory
2017: Richard Thaler − limited rationality, social preferences
12 / 47
Lecture 1: Introduction
Part 1: game theory
“Introduction” / Tour of game theory
Non-cooperative game theory
No binding contracts can be
written
Players are individuals
Main solution concepts:
Nash equilibrium
Strong equilibrium
Cooperative game theory
Binding contract can be written
Players are individuals and
coalitions of individuals
Main solution concepts:
Core
Shapley value
13 / 47
Lecture 1: Introduction
Noncooperative game theory
John Nash (1928-2015)
14 / 47
Lecture 1: Introduction
A noncooperative game (normal-form)
players: N = {1, 2, . . . , n} (finite)
actions/strategies: (each player chooses si from his own finite strategyset; Si for each i ∈ N)
resulting strategy combination: s = (s1, . . . , sn) ∈ (Si)i∈N
payoffs: ui = ui(s)payoffs resulting from the outcome of the game determined by s
15 / 47
Lecture 1: Introduction
Some 2-player examples
Prisoner’s dilemma − social dilemma, tragedy of the commons,free-riding
Conflict between individual and collective incentives
Harmony − aligned incentives
No conflict between individual and collective incentives
Battle of the Sexes − coordination
Conflict and alignment of individual and collective incentives
Hawk dove/Snowdrift − anti-coordination
Conflict and alignment of individual and collective incentives
Matching pennies − zero-sum, rock-paper-scissor
Conflict of individual incentives
16 / 47
Lecture 1: Introduction
Player 2Heads Tails
Player 1Heads 1,-1 -1,1Tails -1,1 1,-1
Matching pennies
17 / 47
Lecture 1: Introduction
Confess Stay quietA A
Confess-6 -10
B -6 0
Stay quiet0 -2
B -10 -2
Prisoner’s dilemma
18 / 47
Lecture 1: Introduction
WOMANBoxing Shopping
MANBoxing 2,1 0,0
Shopping 0,0 1,2
Battle of the sexes
19 / 47
Lecture 1: Introduction
Player 2Hawk Dove
Player 1Hawk -2,-2 4,0Dove 0,4 2,2
Hawk-Dove game
20 / 47
Lecture 1: Introduction
Company BCooperate Not Cooperate
Company ACooperate 9,9 4,7
Not Cooperate 7,4 3,3
Harmony game
21 / 47
Lecture 1: Introduction
Equilibrium
Equilibrium/solution concept:
An equilibrium/solution is a rule that maps the structure of a game into
an equilibrium set of strategies s∗.
22 / 47
Lecture 1: Introduction
Nash Equilibrium
Definition: Best-response
Player i’s best-response (or, reply) to the strategies s−i played by all
others is the strategy s∗i ∈ Si such that
ui(s∗i , s−i) � ui(s′i, s−i) ∀s′i ∈ Si and s′i �= s∗i
Definition: (Pure-strategy) Nash equilibrium
All strategies are mutual best responses:
ui(s∗i , s−i) � ui(s′i, s−i) ∀s′i ∈ Si and s′i �= s∗i
23 / 47
Lecture 1: Introduction
Confess Stay quietA A
Confess-6 -10
B -6 0
Stay quiet0 -2
B -10 -2
Prisoner’s dilemma: both players confess (defect)
24 / 47
Lecture 1: Introduction
WOMANBoxing Shopping
MANBoxing 2,1 0,0
Shopping 0,0 1,2
Battle of the sexes: coordinate on either option
25 / 47
Lecture 1: Introduction
Player 2Heads Tails
Player 1Heads 1,-1 -1,1Tails -1,1 1,-1
Matching pennies: none (in pure strategies)
26 / 47
Lecture 1: Introduction
Player 2Hawk Dove
Player 1Hawk -2,-2 4,0Dove 0,4 2,2
Hawk-dove: either of the two hawk-dove outcomes
27 / 47
Lecture 1: Introduction
Company BCooperate Not Cooperate
Company ACooperate 9,9 4,7
Not Cooperate 7,4 3,3
Harmony: both cooperate
28 / 47
Lecture 1: Introduction
Pure-strategy N.E. for our 2-player examples
Prisoner’s dilemma − social dilemma
Unique NE − socially undesirable outcome
Harmony − aligned incentives
Unique NE − socially desirable outcome
Battle of the Sexes − coordination
Two NE − both Pareto-optimal
Hawk dove/Snowdrift − anti-coordination
Two NE − Pareto-optimal, but perhaps Dove-Dove “better”
Matching pennies − zero-sum, rock-paper-scissor
No (pure-strategy) NE
29 / 47
Lecture 1: Introduction
How about our initial game
Remember the rules were:
1 Choose a number between 0 and 100
2 The player with the number closest to half the average of all submitted
numbers wins
What is the Nash Equilibrium?
30 / 47
Lecture 1: Introduction
0
31 / 47
Lecture 1: Introduction
Finally, let’s play again!
You remember the game:
1 Choose a number between 0 and 100
2 The player with the number that is closest to half the average
43 / 47
This is what others did the 2nd time:
Lecture 1: Introduction
THANKS EVERYBODY
47 / 47