Title: Game Theory and Math Economics: A TCS Introduction
1Game Theory and Math EconomicsA TCS
Introduction
- Christos H. Papadimitriou
- UC Berkeley
- www.cs.berkeley.edu/christos
2Sources
- Osborne and Rubinstein A Course in Game Theory,
MIT, 1994 - Mas-Colell, Whinston, and Greene
- Microeconomic Theory, Oxford, 1995
- these proceedings survey
- http//www.cs.berkeley.edu/christos/games/cs294.h
tml and /focs01.ppt
3- Goal of TCS (1950-2000)
- Develop a mathematical understanding of the
capabilities and limitations of the von Neumann
computer and its software the dominant and most
novel computational artifacts of that time - (Mathematical tools combinatorics, logic)
-
- What should Theorys goals be today?
4(No Transcript)
5The Internet
- built, operated and used by a multitude of
diverse economic interests - theoretical understanding urgently needed
- tools mathematical economics and game theory
6Game Theory
strategies
strategies
3,-2
payoffs
(NB also, many players)
7matching pennies
prisoners dilemma
e.g.
1,-1 -1,1
-1,1 1,-1
3,3 0,4
4,0 1,1
auction
chicken
1 n
1 . . n
0,0 0,1
1,0 -1,-1
0, v y
u x, 0
8concepts of rationality
- undominated strategy
- (problem too weak)
- (weakly) dominating srategy (alias duh?)
- (problem too strong, rarely exists)
- Nash equilibrium (or double best response)
- (problem may not exist)
- randomized Nash equilibrium
- Theorem Nash 1952 Always exists.
. . .
9- if a digraph with all in-degrees ?1 has a source,
- then it must have a sink
- ? Sperners Lemma
- ? Brouwers fixpoint Theorem
- (? Kakutanis Theorem ? market equilibrium)
- Nashs Theorem
- min-max theorem for zero-sum games
- linear programming duality
?
? P
10Sperners Lemma Any legal coloring of the
triangulated simplex has a trichromatic
triangle Proof
!
11Sperner ? Brouwer
- Brouwers Theorem Any continuous function from
the simplex to itself has a fixpoint. - Sketch Triangulate the simplex
- Color vertices according to which direction they
are mapped - Sperners Lemma means that there is a triangle
that has no clear direction - Sequence of finer and finer triangulations,
convergent subsequence of the centers of Sperner
triangles, QED
12Brouwer ? Nash
- For any pair of mixed strategies x,y
(distributions over the strategies) define - ?(x,y) (x, y), where x maximizes
- payoff1(x,y) - x x2,
- and similarly for y. Any Brouwer fixpoint is
now a Nash equilibrium
13Nash ? von Neumann
- If game is zero-sum, then double best response is
the same as max-min pair
14The critique of mixed Nash
- Is it really rational to randomize?
- (cf bluffing in poker, IRS audits)
- If (x,y) is a Nash equilibrium, then any y with
the same support is as good as y. - Convergence/learning results mixed
- There may be too many Nash equilibria
15is it in P?
16The price of anarchy
cost of worst Nash equilibrium
Koutsoupias and P, 1998
socially optimum cost
routing in networks
2 Roughgarden and Tardos, 2000
Also Spirakis and Mavronikolas 01, Roughgarden
01, Koutsoupias and Spirakis 01
The price of the Internet architecture?
17- More problems Nash equilibria may be
politically incorrect Prisoners dilemma - Repeated prisoners dilemma?
- Herb Simon (1969) Bounded Rationality
- the implicit assumption that reasoning and
- computation are infinitely cheap
- is often at the root of negative results in
Economics - Idea Repeated prisoners dilemma played by
memory-limited players (e.g., automata)?
18d
d
c
d
c
d
c
d
c
c, d
c
punish once
tit-for-tat
d
d
c
d
c
d
c,d
c
c
c
d
punish forever
switch on d
Theorem These are the only undominated 2-state
strategies
19how about f(n)-g(n) equilibria?
- Theorem PY94
- (d - d)n
- complicated N.e.
- with payoffs 3 - ?
- tit-for-tat
g(n)
2n
n
f(n)
2n
n
20mechanism design(or inverse game theory)
- agents have utilities but these utilities are
known only to them - game designer prefers certain outcomes depending
on players utilities - designed game (mechanism) has designers goals as
dominating strategies
21mechanism design (math)
- n players, set K of outcomes, for each player i a
possible set Ui of utilities of the form u K ?
R - designer preferences P U1 ? ? Un ? 2K
- mechanism strategy spaces Si, plus a mapping G
S1 ? ? Sn ? K
22Theorem (The Revelation Principle) If there is
a mechanism, then there is one in which all
agents truthfully reveal their secret
utilities. Proof common-sense simulation
Theorem (Gibbard-Satterthwaite) If the sets of
possible utilities are too rich, then only
dictatorial Ps have mechanisms. Proof Arrows
Impossibility Theorem
23- but if we allow mechanisms that use Nash
equilibria instead of dominance, then almost
anything is implementable - but these mechanisms are extremely complex and
artificial - (TCS critique would be welcome here)
24- but if outcomes in K include payments (K K0 ?
Rn ) and utilities are quasilinear (utility of
core outcome plus payment) and designer prefers
to optimize the sum of core utilities, then the
Vickrey-Clarke-Groves mechanism works
25e.g., Vickrey auction
- sealed-highest-bid auction encourages gaming and
speculation - Vickrey auction Highest bidder wins,
- pays second-highest bid
- Theorem Vickrey auction is a truthful
mechanism. - Theorem It maximizes social benefit and
auctioneer expected revenue.
26e.g., shortest path auction
3
6
5
4
t
s
6
10
3
11
pay e its declared cost c(e), plus a bonus equal
to dist(s,t)c(e) ?- dist(s,t)
27- Theorem Resulting mechanism is truthful and
maximizes social benefit - Theorem Suri Hershberger 01 Payments can
be computed by one shortest path computation.
28e.g., 2-processor scheduling Nisan and Ronen
1998
- two players/processors, n tasks, each with a
different execution time on each processor - each execution time is known only to the
appropriate processor - designer wants to minimize makespan
- ( maximum completion time)
- each processor wants to minimize its own
completion time
29- Idea Allocate each task to the most efficient
processor (i.e., minimize total work). Pay each
processor for each task allocated to it an amount
equal to the time required for it at the other
processor -
- Fact Truthful and 2-approximate
30Theorem (Nisan-Ronen) No mechanism can achieve
ratio better than 2
- Sketch By revelation, such a mechanism would be
truthful. - wlog, Processor 1 chooses between proposals of
the form (partition, payment), where the payment
depends only on the partition and Processor 2s
declarations
31Theorem (Nisan-Ronen, continued)
- Suppose all task lengths are 1, and Processor 1
chooses a partition and a payment - If we change the 1-lengths in the partition to ?
and all others to 1 ?, it is not hard to see
that the proposals will remain the same, and
Processor 1 will choose the same one - But this is 2-suboptimal, QED
- Also k processors, randomized 7/4 algorithm.
32e.g., pricing multicasts Feigenbaum, P.,
Shenker, STOC2000
52
30
costs
21
21
40
70
11, 10, 9, 9
14, 8
9, 5, 5, 3
32
23, 17, 14, 9
17, 10
utilities of agents in the node
(u the intrinsic value of the information to
agent i, known only to agent i)
i
33- We wish to design a protocol that will result
- in the computation of
- x ( 0 or 1, will i get it?)
- v (how much will i pay? (0 if x 0) )
- protocol must obey a set of desiderata
i
i
34- 0 ? v ? u,
- lim x 1
- strategy proofness (w u ? x ? v )
- w (u u u ) ? w (u u'u )
-
- welfare maximization
- ? ui xi cT max
-
- marginal cost mechanism
i
i
i
u ??
i
def
i
i
i
i
i
i
i
1
n
1
i
n
- budget balance
- ? v c ( T x)
- Shapley mechanism
i
i
35But
In the context of the Internet, there is another
desideratum Tractability the protocol should
require few (constant? logarithmic?) messages
per link. This new requirement changes
drastically the space of available solutions.
36- 0 ? v ? u
- lim x 1
- strategy proofness (w u ? x ? v )
- w (u u u ) ? w (u u'u )
-
- welfare maximization
- ? w max
-
- marginal cost mechanism
i
i
i
u ??
i
def
i
i
i
i
i
i
i
1
n
1
i
n
- budget balance
- ? v c ( T x)
- Shapley mechanism
i
i
37Bottom-up phase
W ? u ? W ? c, if gt 0 0
otherwise
i
j
c
W
1
W
3
W
2
38Top-down phase
A
c
D min A, W
D
D
v max 0, u ? D
i
i
Theorem The marginal cost mechanism is
tractable.
39Theorem The Shapley value mechanism is
intractable. Model Nodes are linear decision
trees, and they exchange messages that are linear
combinations of the us and cs
c
1
It reduces to checking whether Au gt Bc by two
sites, one of which knows u and the other c,
where A, B are nonsingular
c
2
agents drop out one-by-one
c
n
u lt u lt lt u
1
2
n
40Algorithmic Mechanism Design
- central problem
- few results outside social welfare maximization
framework (n.b.Archer and Tardos 01) - VCG mechanism often breaks the bank
- approximation rarely a remedy (n.b.Nisan and
Ronen 99, Jain and Vazirani 01) - wide open, radical departure needed
41algorithmic aspects of auctions
- Optimal auction design Ronen 01
- Combinatorial auctions Nisan 00
- Auctions for digital goods
- On-line auctions
- Communication complexity of combinatorial
auctions Nisan 01
42coalitional games
Game with players in n v (S) the maximum
total payoff of all players in S,
under worst case play by n
S How to split v (n) fairly?
43first idea the core
A vector (x1, x2,, xn) with ?i x i v(n) is
in the core if for all S we have xS ? v(S)
Problem It is often empty
44second idea the Shapley value
xi E?(vj ?(j) ? ?(i) - vj ?(j) lt ?(i))
e.g., power in the UN Security Council splitting
the cost of a trip
Theorem Shapley The Shapley value is the only
allocation that satisfies Shapleys axioms.
45third idea bargaining setfourth idea
nucleolus ...seventeenth idea the von
Neumann-Morgenstern solution
Deng and P. 1990 complexity-theoretic critique
of solution concepts
46some thoughts on privacy
- also an economic problem
- surrendering private information is either good
or bad for you - personal information is intellectual property
controlled by others, often bearing negative
royalty - selling mailing lists vs. selling aggregate
information false dilemma - Proposal evaluate the individuals contribution
when using personal data for decision-making
47e.g., marketing survey Kleinberg, Raghavan, P
2001
likes
- companys utility is proportional to the
majority - customers utility is 1 if in the majority
- how should all participants be compensated?
customers
possible versions of product
48the internet game
3, 2
capacity of the internal network to carry
traffic (edges have ? capacity)
1, 1
2, 0
1, 4
5, 9
3, 1
intensity of traffic to/from this
node, distributed to other nodes
proportionately to their intensity
3, 6
2, 2
7, 4
3, 1
49vS value of total flow that can be
handled by the subgraph induced by S
- Compute the Shapley flow
- Find a flow in the core
- Under what circumstances is the core
- nonempty? Contains all maximal flows?
50Game Theory and Math Economics
- Deep and elegant
- Different
- Exquisite interaction with TCS
- Relevant to the Internet
- Wide open algorithmic aspects
- Mathematical tools of choice
- for the new TCS