Namespaces
Variants
Actions

Difference between pages "Stability in game theory" and "Bers space"

From Encyclopedia of Mathematics
(Difference between pages)
Jump to: navigation, search
(Importing text file)
 
m (tex encoded by computer)
 
Line 1: Line 1:
A principle reflecting directly or indirectly the idea of stability of a situation (or of a set of situations). One singles out the following basic concepts of stability.
+
<!--
 +
b1103801.png
 +
$#A+1 = 124 n = 0
 +
$#C+1 = 124 : ~/encyclopedia/old_files/data/B110/B.1100380 Bers space
 +
Automatically converted into TeX, above some diagnostics.
 +
Please remove this comment and the {{TEX|auto}} line below,
 +
if TeX found to be correct.
 +
-->
  
1) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869702.png" />-stability, cf. [[Coalitional game|Coalitional game]].
+
{{TEX|auto}}
 +
{{TEX|done}}
  
2) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869704.png" />-stability. An optimality principle in a [[Cooperative game|cooperative game]], connected with the concept of stability of pairs, consisting of a partition of the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869705.png" /> of players into coalitions and allocations relative to the formation of new coalitions. A partition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869706.png" /> of the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869707.png" /> of players is called a coalition structure. Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869708.png" /> be a cooperative game and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s0869709.png" /> a function associating with every coalition structure <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697010.png" /> a set of coalitions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697011.png" />. A pair <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697012.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697013.png" /> is an allocation, is called <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697014.png" />-stable if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697015.png" /> for all <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697016.png" /> and if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697017.png" /> when <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697018.png" />.
+
A complex [[Banach space|Banach space]] of holomorphic automorphic forms introduced by L. Bers (1961). Let  $  D $
 +
be an open set of the Riemann sphere  $  {\widehat{\mathbf C}  } = \mathbf C \cup \{ \infty \} $
 +
whose boundary consists of more than two points. Then  $  D $
 +
carries a unique complete conformal metric  $  \lambda ( z )  | {dz } | $
 +
on  $  D $
 +
with curvature  $  - 4 $,
 +
known as the hyperbolic metric on  $  D $.  
 +
Let $  G $
 +
be a properly discontinuous group of conformal mappings of  $  D $
 +
onto itself (cf. also [[Kleinian group|Kleinian group]]; [[Conformal mapping|Conformal mapping]]). Typical examples of  $  G $
 +
are Kleinian groups (cf. also [[Kleinian group|Kleinian group]]), that is, a group of Möbius transformations (cf. also [[Fractional-linear mapping|Fractional-linear mapping]]) of  $  {\widehat{\mathbf C}  } $
 +
acting properly discontinuously on an open set of $  {\widehat{\mathbf C}  } $.  
 +
By the conformal invariance, the hyperbolic area measure  $  \lambda ( z )  ^ {2}  dx  dy $(
 +
$  z = x + iy $)
 +
on  $  D $
 +
is projected to an area measure  $  d \mu $
 +
on the orbit space  $  D/G $.  
 +
In other words, let  $  d \mu ( w ) = \lambda ( z )  ^ {2}  dx  dy $,
 +
$  w = \pi ( z ) $,
 +
where  $  \pi : D \rightarrow {D/G } $
 +
is the natural projection.
  
3) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697020.png" />-stability. A special case of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697021.png" />-stability, when for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697022.png" /> a set of coalitions is chosen, each of which differs from any element of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697023.png" /> by not more than <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697024.png" /> players.
+
Fix an integer  $  q \geq 2 $.  
 +
A [[Holomorphic function|holomorphic function]]  $  \varphi $
 +
on  $  D $
 +
is called an automorphic form of weight  $  - 2q $
 +
for $  G $
 +
if  $  ( \varphi \circ g ) \cdot ( g  ^  \prime  )  ^ {q} = \varphi $
 +
for all  $  g \in G $.  
 +
Then  $  \lambda ^ {- q } | \varphi | $
 +
is invariant under the action of  $  G $
 +
and hence may be considered as a function on  $  D/G $.  
 +
The Bers space  $  A _ {q}  ^ {p} ( D,G ) $,
 +
where  $  1 \leq  p \leq  \infty $,
 +
is the complex Banach space of holomorphic automorphic forms  $  \varphi $
 +
of weight  $  - 2q $
 +
on  $  D $
 +
for  $  G $
 +
such that the function  $  \lambda ^ {- q } | \varphi | $
 +
on  $  D/G $
 +
belongs to the space  $  L _ {p} $
 +
with respect to the measure  $  \mu $.  
 +
The norm in  $  A _ {q}  ^ {p} ( D,G ) $
 +
is thus given by
  
4) <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697026.png" />-stability. An optimality principle in the theory of cooperative games which formalizes the intuitive notion of stability of formation of coalitions and allocation of values <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697027.png" /> of a characteristic function <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697028.png" /> defined on the set of coalitions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697029.png" /> relative to the possible threat of one coalition against the others. A pair <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697030.png" />, where <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697031.png" /> is a vector satisfying the condition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697032.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697033.png" />, while <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697034.png" /> is a coalition structure, is called a configuration. A configuration is said to be individually rational if <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697035.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697036.png" />. A configuration <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697037.png" /> is called coalitionally rational if the vector <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697038.png" /> satisfies <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697039.png" /> for any coalition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697040.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697041.png" />. In case <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697042.png" />, in particular when <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697043.png" />, for every individually rational configuration <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697044.png" /> the vector <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697045.png" /> is an allocation.
+
$$
 +
\left \| \varphi \right \| = ( {\int\limits \int\limits } _ {D/G }  {\lambda ^ {- pq } \left | \varphi \right |  ^ {p} }  {d \mu } ) ^ {1/p }
 +
$$
  
The set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697046.png" /> is called the set of partners of a coalition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697047.png" /> in a coalition structure <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697048.png" />. Let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697049.png" /> be a coalitionally rational configuration and let <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697050.png" /> be disjoint coalitions. A coalitionally rational configuration <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697051.png" /> satisfying the conditions
+
if  $  1 \leq  p < \infty $,
 +
and
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697052.png" /></td> </tr></table>
+
$$
 +
\left \| \varphi \right \| = \sup  _ {D/G } \lambda ^ {- q } \left | \varphi \right |
 +
$$
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697053.png" /></td> </tr></table>
+
if  $  p = \infty $.
 +
Automorphic forms in  $  A _ {q}  ^ {p} ( D,G ) $
 +
are said to be  $  p $-
 +
integrable if  $  1 \leq  p < \infty $,
 +
and bounded if  $  p = \infty $.  
 +
When  $  G $
 +
is trivial,  $  A _ {q}  ^ {p} ( D,G ) $
 +
is abbreviated to  $  A _ {q}  ^ {p} ( D ) $.  
 +
Note that  $  A _ {q}  ^  \infty  ( D,G ) $
 +
is isometrically embedded as a subspace of  $  A _ {q}  ^  \infty  ( D ) $.
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697054.png" /></td> </tr></table>
+
==Some properties of Bers spaces.==
  
is called a threat of a coalition <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697055.png" /> against <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697056.png" />. By a counter-threat of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697057.png" /> against <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697058.png" /> one understands a coalitionally rational configuration <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697059.png" /> satisfying the conditions
+
1) Let  $  {1 / p } + {1 / {p  ^  \prime  } } = 1 $.  
 +
The Petersson scalar product of $  \varphi \in A _ {q}  ^ {p} ( D,G ) $
 +
and  $  \psi \in A _ {q} ^ {p  ^  \prime  } ( D,G ) $
 +
is defined by
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697060.png" /></td> </tr></table>
+
$$
 +
\left ( \varphi , \psi \right ) = {\int\limits \int\limits } _ {D/G }  {\lambda ^ {- 2q } \varphi {\overline \psi \; } }  {d \mu } .
 +
$$
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697061.png" /></td> </tr></table>
+
If  $  1 \leq  p < \infty $,
 +
then the Petersson scalar product establishes an anti-linear isomorphism of  $  A _ {q} ^ {p  ^  \prime  } ( D,G ) $
 +
onto the dual space of  $  A _ {q}  ^ {p} ( D,G ) $,
 +
whose operator norm is between  $  ( q - 1 ) ( 2q - 1 ) ^ {- 1 } $
 +
and  $  1 $.
  
<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697062.png" /></td> </tr></table>
+
2) The Poincaré (theta-) series of a holomorphic function  $  f $
 +
on  $  D $
 +
is defined by
  
A coalitionally rational configuration <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697063.png" /> is called <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697065.png" />-stable if for any pair of disjoint coalitions <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697066.png" /> and for every threat of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697067.png" /> against <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697068.png" /> there is a counter-threat of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697069.png" /> against <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697070.png" />. The set of all <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697071.png" />-stable configurations for a coalition structure <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697072.png" /> is called the <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697074.png" />-stable set and is denoted by <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697075.png" /> or <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697076.png" />. In the case <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697077.png" />, the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697078.png" /> contains the core (cf. [[Core in the theory of games|Core in the theory of games]]) of the cooperative game <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697079.png" />. The set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697080.png" /> often turns out to be empty, and therefore one considers further the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697081.png" /> which is defined analogously to <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697082.png" />, with the following changes: one considers not only coalitionally rational configurations, but all individually rational configurations admitting only threats and counter-threats among one-element coalitions, i.e. between individual players. It can be proved that the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697083.png" /> is non-empty for any coalition structure. The set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697084.png" /> for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697085.png" /> contains the <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697086.png" />-kernel and coincides with it and the core for a [[Convex game|convex game]] <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697087.png" />.
+
$$
 +
\Theta f = \sum _ {g \in G } ( f \circ g ) \cdot ( g  ^  \prime  ) ^ {q}
 +
$$
  
The concepts of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697088.png" />-stability and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697089.png" />-stability have a natural generalization to cooperative games without side payments. It is known that in this case the set <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697090.png" /> may be empty; there are certain conditions for <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697091.png" /> to be non-empty.
+
whenever the right-hand side converges absolutely and uniformly on compact subsets of  $  D $(
 +
cf. [[Absolutely convergent series|Absolutely convergent series]]; [[Uniform convergence|Uniform convergence]]). Then  $  \Theta f $
 +
is an automorphic form of weight  $  - 2q $
 +
on  $  D $
 +
for  $  G $.
 +
Moreover,  $  \Theta $
 +
gives a continuous linear mapping of  $  A _ {q}  ^ {1} ( D ) $
 +
onto  $  A _ {q}  ^ {1} ( D,G ) $
 +
of norm at most  $  1 $.
 +
For every  $  \varphi \in A _ {q}  ^ {p} ( D,G ) $
 +
there exists an  $  f \in A _ {q}  ^ {p} ( D ) $
 +
with  $  \| f \| \leq  ( 2q - 1 ) ( q - 1 ) ^ {- 1 } \| \varphi \| $
 +
such that  $  \varphi = \Theta f $.
 +
 
 +
3) Let  $  B $
 +
be the set of branch points of the natural projection  $  \pi $.
 +
Assume that: i)  $  D/G $
 +
is obtained from a (connected) closed [[Riemann surface|Riemann surface]] of genus  $  g $
 +
by deleting precisely  $  m $
 +
points; and ii)  $  \pi ( B ) $
 +
consists of exactly  $  n $
 +
points  $  p _ {1} \dots p _ {n} $(
 +
possibly,  $  m = 0 $
 +
or  $  n =0 $).  
 +
For each  $  k = 1 \dots n $,
 +
let  $  \nu _ {k} $
 +
be the common multiplicity of  $  \pi $
 +
at points of  $  \pi ^ {- 1 } ( p _ {k} ) $.  
 +
Then  $  A _ {q}  ^ {p} ( D,G ) = A _ {q}  ^  \infty  ( D,G ) $
 +
for  $  1 \leq  p \leq  \infty $
 +
and
 +
 
 +
$$
 +
{ \mathop{\rm dim} } A _ {q}  ^  \infty  ( D,G ) =
 +
$$
 +
 
 +
$$
 +
=  
 +
( 2q - 1 ) ( g - 1 ) + ( q - 1 ) m + \sum _ {k =1 } ^ { n }  \left [ q \left ( 1 - {
 +
\frac{1}{\nu _ {k} }
 +
} \right ) \right ] ,
 +
$$
 +
 
 +
where  $  [ x ] $
 +
denotes the largest integer that does not exceed  $  x $.
 +
 
 +
4) Consider the particular case where  $  D $
 +
is the unit disc. Then  $  G $
 +
is a [[Fuchsian group|Fuchsian group]] and  $  \lambda ( z ) = ( 1 - | z |  ^ {2} ) ^ {- 1 } $.  
 +
It had been conjectured that $  A _ {q}  ^ {1} ( D,G ) \subset  A _ {q}  ^  \infty  ( D,G ) $
 +
for any  $  G $,
 +
until Ch. Pommerenke [[#References|[a6]]] constructed a counterexample. In [[#References|[a5]]] D. Niebur and M. Sheingorn characterized the Fuchsian groups  $  G $
 +
for which the inclusion relation holds. In particular, if  $  G $
 +
is finitely generated, then  $  A _ {q}  ^ {1} ( D,G ) \subset  A _ {q}  ^  \infty  ( D,G ) $.
 +
 
 +
5) Let  $  G $
 +
be a Fuchsian group acting on the unit disc  $  D $.  
 +
It also preserves  $  D  ^ {*} = {\widehat{\mathbf C}  } \setminus  {\overline{D}\; } $,
 +
the outside of the unit circle. If  $  f $
 +
is conformal on  $  D  ^ {*} $
 +
and can be extended to a [[Quasi-conformal mapping|quasi-conformal mapping]] of  $  {\widehat{\mathbf C}  } $
 +
onto itself such that  $  f \circ g \circ f ^ {- 1 } $
 +
is a Möbius transformation for each  $  g \in G $,
 +
then its [[Schwarzian derivative|Schwarzian derivative]]
 +
 
 +
$$
 +
Sf = {
 +
\frac{f ^ {\prime \prime \prime } }{f  ^  \prime  }
 +
} - {
 +
\frac{3}{2}
 +
} \left ( {
 +
\frac{f ^ {\prime \prime } }{f  ^  \prime  }
 +
} \right )  ^ {2}
 +
$$
 +
 
 +
belongs to  $  A _ {2}  ^  \infty  ( D  ^ {*} ,G ) $
 +
with  $  \| {Sf } \| \leq  6 $.  
 +
Moreover, the set of all such Schwarzian derivatives constitutes a bounded domain in  $  A _ {2}  ^  \infty  ( D  ^ {*} ,G ) $
 +
including the open ball of radius  $  2 $
 +
centred at the origin. This domain can be regarded as a realization of the [[Teichmüller space|Teichmüller space]]  $  T ( G ) $
 +
of  $  G $,
 +
and the injection of  $  T ( G ) $
 +
into  $  A _ {2}  ^  \infty  ( D  ^ {*} ,G ) $
 +
induced by the Schwarzian derivative is referred to as the Bers embedding.
  
 
====References====
 
====References====
<table><TR><TD valign="top">[1]</TD> <TD valign="top">  R.J. Aumann,  M. Maschler,  "The bargaining set for cooperative games" , ''Advances in game theory'' , Princeton Univ. Press pp. 443–476</TD></TR><TR><TD valign="top">[2]</TD> <TD valign="top">  N.N. Vorob'ev,  "The present state of the theory of games" ''Russian Math. Surveys'' , '''25''' :  2  (1970)  pp. 77–150  ''Uspekhi Mat. Nauk'' , '''25''' :  2 (1970) pp. 81–140</TD></TR><TR><TD valign="top">[3]</TD> <TD valign="top">  R.D. Luce,  , ''Mathematical Models of Human Behaviour'' , Stanford (1955)  pp. 32–44</TD></TR><TR><TD valign="top">[4]</TD> <TD valign="top">  R.D. Luce,  H. Raiffa,  "Games and decisions. Introduction and critical survey" , Wiley  (1957)</TD></TR><TR><TD valign="top">[5]</TD> <TD valign="top">  B. Peleg,  "Existence theorem for the bargaining of <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/s/s086/s086970/s08697092.png" />"  ''Bull. Amer. Math. Soc.'' , '''69'''  (1963)  pp. 109–110</TD></TR><TR><TD valign="top">[6]</TD> <TD valign="top">  B. Peleg,  "Quota games with a continuum of players"  ''Israel J. Math.'' , '''1'''  (1963)  pp. 48–53</TD></TR><TR><TD valign="top">[7]</TD> <TD valign="top">  G. Owen,  "The theory of games" , Acad. Press  (1982)</TD></TR></table>
+
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  I. Kra,  "Automorphic forms and Kleinian groups" , Benjamin (1972)</TD></TR><TR><TD valign="top">[a2]</TD> <TD valign="top">  J. Lehner,  "Discontinuous groups and automorphic functions" , Amer. Math. Soc.  (1964)</TD></TR><TR><TD valign="top">[a3]</TD> <TD valign="top">  J. Lehner,   "Automorphic forms" W.J. Harvey (ed.) , ''Discrete Groups and Automorphic Functions'' , Acad. Press (1977)  pp. 73–120</TD></TR><TR><TD valign="top">[a4]</TD> <TD valign="top">  S. Nag,  "The complex analytic theory of Teichmüller spaces" , Wiley  (1988)</TD></TR><TR><TD valign="top">[a5]</TD> <TD valign="top">  D. Niebur,  M. Sheingorn,  "Characterization of Fuchsian groups whose integrable forms are bounded"  ''Ann. of Math.'' , '''106'''  (1977)  pp. 239–258</TD></TR><TR><TD valign="top">[a6]</TD> <TD valign="top">  Ch. Pommerenke,  "On inclusion relations for spaces of automorphic formsW.E. Kirwan (ed.)  L. Zalcman (ed.) , ''Advances in Complex Function Theory'' , ''Lecture Notes in Mathematics'' , '''505''' , Springer (1976)  pp. 92–100</TD></TR></table>

Latest revision as of 10:58, 29 May 2020


A complex Banach space of holomorphic automorphic forms introduced by L. Bers (1961). Let $ D $ be an open set of the Riemann sphere $ {\widehat{\mathbf C} } = \mathbf C \cup \{ \infty \} $ whose boundary consists of more than two points. Then $ D $ carries a unique complete conformal metric $ \lambda ( z ) | {dz } | $ on $ D $ with curvature $ - 4 $, known as the hyperbolic metric on $ D $. Let $ G $ be a properly discontinuous group of conformal mappings of $ D $ onto itself (cf. also Kleinian group; Conformal mapping). Typical examples of $ G $ are Kleinian groups (cf. also Kleinian group), that is, a group of Möbius transformations (cf. also Fractional-linear mapping) of $ {\widehat{\mathbf C} } $ acting properly discontinuously on an open set of $ {\widehat{\mathbf C} } $. By the conformal invariance, the hyperbolic area measure $ \lambda ( z ) ^ {2} dx dy $( $ z = x + iy $) on $ D $ is projected to an area measure $ d \mu $ on the orbit space $ D/G $. In other words, let $ d \mu ( w ) = \lambda ( z ) ^ {2} dx dy $, $ w = \pi ( z ) $, where $ \pi : D \rightarrow {D/G } $ is the natural projection.

Fix an integer $ q \geq 2 $. A holomorphic function $ \varphi $ on $ D $ is called an automorphic form of weight $ - 2q $ for $ G $ if $ ( \varphi \circ g ) \cdot ( g ^ \prime ) ^ {q} = \varphi $ for all $ g \in G $. Then $ \lambda ^ {- q } | \varphi | $ is invariant under the action of $ G $ and hence may be considered as a function on $ D/G $. The Bers space $ A _ {q} ^ {p} ( D,G ) $, where $ 1 \leq p \leq \infty $, is the complex Banach space of holomorphic automorphic forms $ \varphi $ of weight $ - 2q $ on $ D $ for $ G $ such that the function $ \lambda ^ {- q } | \varphi | $ on $ D/G $ belongs to the space $ L _ {p} $ with respect to the measure $ \mu $. The norm in $ A _ {q} ^ {p} ( D,G ) $ is thus given by

$$ \left \| \varphi \right \| = ( {\int\limits \int\limits } _ {D/G } {\lambda ^ {- pq } \left | \varphi \right | ^ {p} } {d \mu } ) ^ {1/p } $$

if $ 1 \leq p < \infty $, and

$$ \left \| \varphi \right \| = \sup _ {D/G } \lambda ^ {- q } \left | \varphi \right | $$

if $ p = \infty $. Automorphic forms in $ A _ {q} ^ {p} ( D,G ) $ are said to be $ p $- integrable if $ 1 \leq p < \infty $, and bounded if $ p = \infty $. When $ G $ is trivial, $ A _ {q} ^ {p} ( D,G ) $ is abbreviated to $ A _ {q} ^ {p} ( D ) $. Note that $ A _ {q} ^ \infty ( D,G ) $ is isometrically embedded as a subspace of $ A _ {q} ^ \infty ( D ) $.

Some properties of Bers spaces.

1) Let $ {1 / p } + {1 / {p ^ \prime } } = 1 $. The Petersson scalar product of $ \varphi \in A _ {q} ^ {p} ( D,G ) $ and $ \psi \in A _ {q} ^ {p ^ \prime } ( D,G ) $ is defined by

$$ \left ( \varphi , \psi \right ) = {\int\limits \int\limits } _ {D/G } {\lambda ^ {- 2q } \varphi {\overline \psi \; } } {d \mu } . $$

If $ 1 \leq p < \infty $, then the Petersson scalar product establishes an anti-linear isomorphism of $ A _ {q} ^ {p ^ \prime } ( D,G ) $ onto the dual space of $ A _ {q} ^ {p} ( D,G ) $, whose operator norm is between $ ( q - 1 ) ( 2q - 1 ) ^ {- 1 } $ and $ 1 $.

2) The Poincaré (theta-) series of a holomorphic function $ f $ on $ D $ is defined by

$$ \Theta f = \sum _ {g \in G } ( f \circ g ) \cdot ( g ^ \prime ) ^ {q} $$

whenever the right-hand side converges absolutely and uniformly on compact subsets of $ D $( cf. Absolutely convergent series; Uniform convergence). Then $ \Theta f $ is an automorphic form of weight $ - 2q $ on $ D $ for $ G $. Moreover, $ \Theta $ gives a continuous linear mapping of $ A _ {q} ^ {1} ( D ) $ onto $ A _ {q} ^ {1} ( D,G ) $ of norm at most $ 1 $. For every $ \varphi \in A _ {q} ^ {p} ( D,G ) $ there exists an $ f \in A _ {q} ^ {p} ( D ) $ with $ \| f \| \leq ( 2q - 1 ) ( q - 1 ) ^ {- 1 } \| \varphi \| $ such that $ \varphi = \Theta f $.

3) Let $ B $ be the set of branch points of the natural projection $ \pi $. Assume that: i) $ D/G $ is obtained from a (connected) closed Riemann surface of genus $ g $ by deleting precisely $ m $ points; and ii) $ \pi ( B ) $ consists of exactly $ n $ points $ p _ {1} \dots p _ {n} $( possibly, $ m = 0 $ or $ n =0 $). For each $ k = 1 \dots n $, let $ \nu _ {k} $ be the common multiplicity of $ \pi $ at points of $ \pi ^ {- 1 } ( p _ {k} ) $. Then $ A _ {q} ^ {p} ( D,G ) = A _ {q} ^ \infty ( D,G ) $ for $ 1 \leq p \leq \infty $ and

$$ { \mathop{\rm dim} } A _ {q} ^ \infty ( D,G ) = $$

$$ = ( 2q - 1 ) ( g - 1 ) + ( q - 1 ) m + \sum _ {k =1 } ^ { n } \left [ q \left ( 1 - { \frac{1}{\nu _ {k} } } \right ) \right ] , $$

where $ [ x ] $ denotes the largest integer that does not exceed $ x $.

4) Consider the particular case where $ D $ is the unit disc. Then $ G $ is a Fuchsian group and $ \lambda ( z ) = ( 1 - | z | ^ {2} ) ^ {- 1 } $. It had been conjectured that $ A _ {q} ^ {1} ( D,G ) \subset A _ {q} ^ \infty ( D,G ) $ for any $ G $, until Ch. Pommerenke [a6] constructed a counterexample. In [a5] D. Niebur and M. Sheingorn characterized the Fuchsian groups $ G $ for which the inclusion relation holds. In particular, if $ G $ is finitely generated, then $ A _ {q} ^ {1} ( D,G ) \subset A _ {q} ^ \infty ( D,G ) $.

5) Let $ G $ be a Fuchsian group acting on the unit disc $ D $. It also preserves $ D ^ {*} = {\widehat{\mathbf C} } \setminus {\overline{D}\; } $, the outside of the unit circle. If $ f $ is conformal on $ D ^ {*} $ and can be extended to a quasi-conformal mapping of $ {\widehat{\mathbf C} } $ onto itself such that $ f \circ g \circ f ^ {- 1 } $ is a Möbius transformation for each $ g \in G $, then its Schwarzian derivative

$$ Sf = { \frac{f ^ {\prime \prime \prime } }{f ^ \prime } } - { \frac{3}{2} } \left ( { \frac{f ^ {\prime \prime } }{f ^ \prime } } \right ) ^ {2} $$

belongs to $ A _ {2} ^ \infty ( D ^ {*} ,G ) $ with $ \| {Sf } \| \leq 6 $. Moreover, the set of all such Schwarzian derivatives constitutes a bounded domain in $ A _ {2} ^ \infty ( D ^ {*} ,G ) $ including the open ball of radius $ 2 $ centred at the origin. This domain can be regarded as a realization of the Teichmüller space $ T ( G ) $ of $ G $, and the injection of $ T ( G ) $ into $ A _ {2} ^ \infty ( D ^ {*} ,G ) $ induced by the Schwarzian derivative is referred to as the Bers embedding.

References

[a1] I. Kra, "Automorphic forms and Kleinian groups" , Benjamin (1972)
[a2] J. Lehner, "Discontinuous groups and automorphic functions" , Amer. Math. Soc. (1964)
[a3] J. Lehner, "Automorphic forms" W.J. Harvey (ed.) , Discrete Groups and Automorphic Functions , Acad. Press (1977) pp. 73–120
[a4] S. Nag, "The complex analytic theory of Teichmüller spaces" , Wiley (1988)
[a5] D. Niebur, M. Sheingorn, "Characterization of Fuchsian groups whose integrable forms are bounded" Ann. of Math. , 106 (1977) pp. 239–258
[a6] Ch. Pommerenke, "On inclusion relations for spaces of automorphic forms" W.E. Kirwan (ed.) L. Zalcman (ed.) , Advances in Complex Function Theory , Lecture Notes in Mathematics , 505 , Springer (1976) pp. 92–100
How to Cite This Entry:
Stability in game theory. Encyclopedia of Mathematics. URL: http://encyclopediaofmath.org/index.php?title=Stability_in_game_theory&oldid=14883
This article was adapted from an original article by A.Ya. Kiruta (originator), which appeared in Encyclopedia of Mathematics - ISBN 1402006098. See original article