HAL Id: inria-00360829
https://hal.inria.fr/inria-00360829
Submitted on 12 Feb 2009
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
To cite this version:
RANDOM FRUITS ON THE ZIELONKA TREE
FLORIANHORN
1
1
CWI,Amsterdam,TheNetherlands E-mailaddress: f.horn wi.nl
Abstra t. Sto hasti games are a natural model for the synthesis of ontrollers on-frontedto adversarialand/or random a tions. Inparti ular,
ω
-regulargamesof innite length anrepresentrea tivesystemswhi harenotexpe tedtorea ha orre tstate,but rathertohandlea ontinuousstreamofevents. One riti alresour einsu happli ations isthememoryusedbythe ontroller. Inthispaper,westudytheamountofmemorythat anbesavedthroughtheuseofrandomisationinstrategies, andpresentmat hingupper andlowerboundsforsto hasti Mullergames.1. Introdu tion
A sto hasti gamearena is a dire ted graph with three kindsof states: Eve's, Adam's and random states. A token ir ulates on this arena: when it is in one of Eve's states, she hooses its next lo ation among the su essors of the urrent state; when it is in one of Adam's states, he hooses its next lo ation; and when it is ina random state, the next lo ation is hosen a ording to a xed probability distribution. The result of playing the game for
ω
moves is an innite path of the graph. A play is winning either for Eve or for Adam, and the winner problem onsists in determining whether one of the players has a winning strategy, from a given initial state. Closely related problems on ern the omputationofwinningstrategies,aswellasdeterminingthenatureofthesestrategies: pure orrandomised,with niteor innitememory. Therehasbeenalonghistoryofusingarenas without random states (2-player arenas) for modelling and synthesising rea tive pro esses [BL69 , PR89℄: Eve represents the ontroller, and Adam theenvironment. Sto hasti (21
2
-player) arenas [deA97℄, withthe addition of random states, an also model un ontrollable a tionsthathappen a ordingtoarandom law,rather thanby hoi e ofana tively hostile environment. The desired behaviour of the system is traditionally represented as anω
-regularwinning ondition,whi hnaturallyexpressesthetemporalspe i ationsandfairness assumptions of transition systems [MP92 ℄. From this point of view, the omplexity of the winningstrategiesisa entralquestion,sin etheyrepresentpossibleimplementationsofthe ontrollers inthe synthesisproblem. In this paper, we fo uson animportant normalform ofω
-regular onditions,namely Muller winning onditions (see [Tho95℄for a survey).1998ACMSubje tClassi ation: D.2.4. ModelChe king(Theory).
Keywordsandphrases: model he king, ontrollersynthesis,sto hasti games,randomisation.
Thisworkwas arriedoutduringthetenureofanERCIM"AlainBensoussan"FellowshipProgramme.
c
Florian Horn
CC
In the ase of 2-player Muller games, a fundamental determina y result of Bü hi and Landweber states that, from any initial state, one of the players has a winning strategy [BL69 ℄. Gurevi handHarrington usedthelatest appearan e re ord (LAR)stru tureof M -Naughton to extend this result to strategies withmemory fa torial inthesize ofthe game [GH82℄. Zielonka renes the LAR onstru tion into a tree, and derives from it an elegant algorithmto omputethewinningregions[Zie98 ℄. AninsightfulanalysisoftheZielonkatree by Dziembowski, Jurdzinski, and Walukiewi z leads to optimal (and asymmetri al) mem-ory bounds for pure (non-randomised) winning strategies [DJW97 ℄. Chatterjee extended these bounds to the ase of pure strategies over 2
1
2
-player arenas [Cha07b℄. However, the lowerboundon memorydoesnotholdforrandomisedstrategies, eveninnon-sto hasti are-nas: Chatterjee, de Alfaro,and Henzingershowthatmemoryless randomised strategiesare enough for to deal with upward- losed winning onditions [CdAH04 ℄. Chatterjee extends thisresultin[Cha07a℄,showingthat onditionswithnon-trivialupward- losedsubsetsadmit randomised strategieswithlessmemory than pure ones.Our ontributions. The memory bounds of [Cha07a ℄ are not tight in general, even for 2-player arenas. We give here mat hing upperand lower boundsfor any Muller ondition
F
,intheform ofa numberr
F
omputed fromthe Zielonka treeofF
:•
if Eve has a winning strategy in a2
1
2
-player game(A, F)
, she has a randomised winning strategy withmemoryr
F
(Theorem 4.2);•
there is a 2-player game(A
F
, F)
where any randomised winning strategy for Eve hasat leastr
F
memory states(Theorem 5.2).Furthermore, the witness arenas we build in the proof of Theorem 5.2 are signi antly smallerthan in[DJW97 ℄, even thoughtheproblemof polynomial arenas remainsopen. Outline of the paper. Se tion 2 re alls the lassi al notions inthe area, whileSe tion 3 presentsformerresultsonmemoryboundsandrandomisedstrategies. Thenexttwose tions present our main results. In Se tion 4, we introdu e the number
r
F
and show that itis an upperbound on thememory needed to win inany21
2
-game(A, F)
. InSe tion 5, we show thatthisboundistight. Finally,inSe tion 6,we hara terisethe lassofMuller onditions thatadmit memorylessrandomised strategies, and show thatfor ea h Muller ondition, at leastone ofthe players annot improve itsmemory throughrandomisation.2. Denitions
We onsider turn-based sto hasti two-player Muller games. We re all here several lassi al notions inthe eld,and refer thereaderto [Tho95,deA97℄ formore details. ProbabilityDistribution. Aprobabilitydistribution
γ
overasetX
isafun tionfromX
to[0, 1]
su h thatP
x∈X
γ(x) = 1
. The set of probability distributions overX
is denoted byD(X)
.Arenas. A 2
1
2
-player arenaA
over a set of oloursC
onsists of a dire ted nite graph(S, T )
,apartition(S
E
, S
A
, S
R
)
ofS
,aprobabilisti transitionfun tionδ : S
R
→ D(S)
su h thatδ(s)(t) > 0 ⇔ (s, t) ∈ T
, anda partial olouring fun tionχ : S ⇀ C
. ThestatesinS
E
(resp.S
A
,S
R
) are Eve's states (resp. Adam's states, random states), and are graphi ally representedas#
's (resp.2
,△
). A 2-player arena isan arenawhereS
R
= ∅
.Aset
U ⊆ S
ofstatesisδ
- losed ifforeveryrandomstateu ∈ U ∩S
R
,(u, t) ∈ T → t ∈ U
. It islive iffor every non-randomstateu ∈ U ∩ (S
E
∪ S
A
)
,there isa statet ∈ U
su h thatPlays and Strategies. Aninnite path, or play, over thearena
A
is an innitesequen eρ = ρ
0
ρ
1
. . .
of states su h that(ρ
i
, ρ
i+1
) ∈ T
for alli ∈
N
. The set of states o urring innitelyofteninaplayρ
isdenoted byInf(ρ) = {s | ∃
∞
i ∈
N, ρ
i
= s}
. WewriteΩ
forthe setof all plays,andΩ
s
for theset ofplays thatstart fromthestates
.A strategy with memory
M
for Eve on the arenaA
is a (possibly innite) transdu erσ = (M, σ
n
, σ
u
)
, whereσ
n
is the next-move fun tion from
(S
E
× M )
toD(S)
andσ
u
is the memory-update fun tion, from
(S × M )
toD(M )
. Noti e that both the move and the update are randomised: strategies whose memory is deterministi are a dierent, less ompa t, model. The strategies for Adam are dened likewise. A strategyσ
is pure if it doesnot userandomisation. It is nite-memory ifM
is a nite set, and memoryless ifM
is a singleton. Noti e that strategies dened in the usual way as fun tions fromS
∗
to
S
an be dened as strategies withinnite memory: theset of memory states isS
∗
and the memory update is
σ
u
(s, w) 7→ ws
.On eastartingstate
s ∈ S
andstrategiesσ ∈ Σ
for bothplayersarexed,theout ome of the game is a random walkρ
σ,τ
s
for whi h the probabilities of events are uniquely xed (an event is a measurable set of paths). For an eventP ∈ Ω
, we denote byP
σ,τ
s
(P )
the probabilitythataplaybelongstoP
ifitstartsfroms
andEveandAdamfollowthestrategiesσ
andτ
.A playis onsistent with
σ
iffor ea h positioni
su hthatw
i
∈ S
E
,P
σ,τ
w
0
(ρ
i+1
= w
i+1
|
ρ
0
= w
0
. . . ρ
i
= w
i
) > 0
. The set of plays onsistent withσ
is denoted byΩ
σ
. Similar notions an be dened for Adam'sstrategies.
Traps andAttra tors. Theattra tor ofEve tothe set
U
,denotedAttr
E
(U )
,isthesetof stateswhereEve anguaranteethatthetokenrea hesthesetU
withapositiveprobability. It isdened indu tively by:Attr
0
E
(U )
=
U
Attr
i+1
E
(U ) = Attr
i
E
(U ) ∪{s ∈ S
E
∪ S
R
, ∃t ∈ Attr
i
E
(U ) | (s, t) ∈ T }
∪{s ∈ S
A
| ∀t, (s, t) ∈ E ⇒ t ∈ Attr
i
E
(U )}
Attr
E
(U )
=
S
i>0
Attr
i
E
(U )
The orresponding attra tor strategy to
U
for Eve is a pure and memoryless strategya
U
su hthat for anystates ∈ S
E
∩ (Attr
E
(U ) \ U )
,s ∈ Attr
i+1
E
(U ) ⇒ a
U
(s) ∈ Attr
i
E
(U )
. The dual notion of trap for Eve denotes a set from where Eve annot es ape, unless Adamallowshertodoso: asetU
isatrapforEveifandonlyif∀s ∈ U ∩ (S
E
∪ S
R
), (s, t) ∈
T ⇒ t ∈ U
and∀s ∈ U ∩ S
A
, ∃t ∈ U, (s, t) ∈ T
. Noti e that a trap is a strong notion the token an neverleave itifAdam does not allowit to do so whilean attra tor is a weak one thetoken an avoidthetarget even ifEveuses theattra torstrategy. Noti e also thata trap(foreitherplayer) isalways a subarena.Winning Conditions. A winning ondition is a subset
Φ
ofΩ
. A playρ
is winning for Eve ifρ ∈ Φ
, and winning for Adam otherwise. We onsiderω
-regular winning onditions formalised as Muller onditions. A Muller ondition is determined by a subsetF
of the power setP(C)
of olours, and Eve wins a play if and only if the set of olours visited innitelyoften belongs toF
:Φ
F
= {ρ ∈ Ω|χ(Inf(ρ)) ∈ F}
. An exampleof Muller game is giveninFigure1(a). Weuseitthroughoutthepapertodes ribevariousnotions andresults. WinningStrategies. Astrategyσ
forEveissurelywinning(orsure)fromastates
forthe winning onditionΦ
ifanyplay onsistent withσ
belongs toΦ
,and almost-surely winning (or almost-sure) if for any strategyτ
for Adam,P
σ,τ
c
d
a
b
a
b
c
F
= {{a, b}, {a, b, c}, {a, b, c, d}}
(a)ThegameG= (
A,
F)
abcd
bcd acd abd
ab
a
b
(b)ZielonkaTreeofF
Figure1: Re urringExample
3. Former results in memory bounds and randomisation
3.1. Pure strategies
Therehasbeenintenseresear hsin ethesixtiesonthenon-sto hasti setting,i.e. pure strategies and 2-player arenas. Bü hi and Landweber showed the determina y of Muller games in[BL69 ℄. Gurevi h and Harrington used theLAR (Latest Appearan e Re ord) of M Naughton to prove their Forgetful Determina y theorem [GH82 ℄, whi h shows that a memory of size
|C|!
issu ient for any gamethat uses only oloursfromC
,even whenthe arenais innite. Thisresult waslater rened byZielonka in[Zie98 ℄, using arepresentation of theMuller onditions astrees:Denition 3.1(ZielonkaTreeofaMuller ondition). TheZielonkaTree
Z
F ,C
ofawinning onditionF ⊆ P(C)
isdened indu tively asfollows:(1) If
C /
∈ F
,thenZ
F ,C
= Z
F ,C
,whereF = P(C) \ F
.(2) If
C ∈ F
, then the root ofZ
F ,C
is labelled withC
. LetC
1
, C
2
, . . . , C
k
be all the maximal sets in{U /
∈ F | U ⊆ C}
. Then we atta h to theroot, asits subtrees, the Zielonka trees ofF ↾ C
i
,i.e. theZ
F ↾C
i
,C
i
,fori = 1 . . . k
.Hen e, the Zielonka tree is atree with nodeslabelled by sets of olours. A node of
Z
F ,C
is an Eve node ifitislabelled witha set fromF
,otherwise itisan Adamnode.A later analysis of this onstru tion by Dziembowski, Jurdzinski and Walukiewi z in [DJW97 ℄led to an optimaland asymmetri al boundon thememory needed bythe players to dene surestrategies:
Denition 3.2 (Number
m
F
ofa Muller ondition). LetF ⊆ P(C)
bea Muller ondition, andZ
F
1
,C
1
, Z
F
2
,C
2
, . . . , Z
F
k
,C
k
be the subtrees atta hed to the root of the treeZ
F ,C
. We dene thenumberm
F
indu tively asfollows:m
F
=
1
ifZ
F ,C
doesnot have anysubtrees,max{m
F
1
, m
F
2
, . . . , m
F
k
}
ifC /
∈ F
(Adamnode),k
X
i=1
m
F
i
ifC ∈ F
(Eve node).Theorem 3.4 ([Cha07b℄). If Eve has an almost-sure strategy in a 2
1
2
-player Muller game withthewinning onditionF
,she hasa purealmost-surewithatmostm
F
memory states.3.2. Memory redu tion through randomisation
Randomisedstrategiesaremoregeneralthanpurestrategies,andinsome ases,theyare alsomore ompa t. In[CdAH04 ℄,arstresultshowedthatupward- losed onditionsadmit memorylessrandomised strategies, while theydon't admit memorylesspurestrategies: Theorem 3.5 ([CdAH04℄). If Eve has an almost-sure strategy in a 2
1
2
-player Muller game withan upward- losed winning ondition,she has a randomised almost-sure strategy.Thisresult was later extendedin [Cha07a ℄,byremoving theleavesatta hed to a node of theZielonkaTreerepresenting anupward- losed sub ondition:
Denition 3.6 ([Cha07a ℄). Let
F ⊆ P(C)
be a Muller ondition, andZ
F
1
,C
1
, Z
F
2
,C
2
, . . . ,
Z
F
k
,C
k
be the subtrees atta hed to the root of the treeZ
F ,C
. We dene the numberm
U
F
indu tively asfollows:m
U
F
=
1
ifZ
F ,C
doesnot have anysubtrees,1
ifF
isupward- losed,max{m
U
F
1
, m
U
F
2
, . . . , m
U
F
k
}
ifC /
∈ F
(Adamnode),k
X
i=1
m
U
F
i
ifC ∈ F
(Eve node).Theorem 3.7 ([Cha07a℄). If Eve has an almost-sure strategy in a 2
1
2
-player Muller game with the winning onditionF
, she has a randomised almost-sure strategy with at mostm
U
F
memory states.
4. Randomised Upper Bound
TheupperboundofTheorem3.7isnottightforall onditions. Forexample,thenumber
m
U
F
ofthe onditionFinFigure1(b)isthree,whilethereisalwaysanalmost-sure strategy withtwo memory states. We present hereyet anothernumber for any Muller ondition
F
, denotedr
F
,thatwe ompute fromthe Zielonka Tree:Denition 4.1 (Number
r
F
of a Muller ondition). LetF ⊆ P(C)
be a Muller ondition, where theroot hask + l
hildren,l
of thembeingleaves. We denote byZ
F
1
,C
1
, Z
F
2
,C
2
, . . . ,
Z
F
k
,C
k
the non-leaves subtrees atta hed to the root ofZ
F ,C
. We dener
F
indu tively as follows:r
F
=
1
ifZ
F ,C
doesnot have anysubtrees,max{1, r
F
1
, r
F
2
, . . . , r
F
k
}
ifC /
∈ F
(Adamnode),k
X
i=1
r
F
i
ifC ∈ F
(Eve node)andl = 0,
k
X
i=1
The rst remark is that if
∅ ∈ F
,r
F
is equal tom
F
: asthe leaves belong to Eve, the fourth ase annoto ur. Intheother ase, theintuitionisthatwemergeleavesiftheyare siblings. For example,the numberr
F
for our re urring example istwo: one for the leaves labelled
bcd
andacd
,and one for theleaveslabelleda
andb
. Thenumberm
F
isfour (one for ea h leaf),and
m
U
F
isthree (onefor the leaves labelled
a
andb
,and one for ea h other leaf). Thisse tion willbedevotedto theproofof Theorem 4.2:Theorem 4.2 (Randomised upper bound). If Eve has an almost-sure strategy in a
2-1
2
player Muller game with the winning ondition
F)
, she has an almost-sure strategy with memoryr
F
.Let
G = (F, A)
be agame dened on thesetof oloursC
su h thatEve winsfrom any initial node. We des ribeinthenextthreesubse tionsare ursive pro edureto ompute an almost-surestrategyforEvewithr
F
memorystatesinea hnon-trivial aseinthedenition ofr
F
. WeusetwolemmasLemmas4.3and4.5thatderivedire tlyfromsimilarresults in [DJW97 ℄ and [Cha07b ℄. The appli ation of these prin iples to the game G in Figure 1 builds a randomised strategy with two memory statesleft and right. Inleft, Eve sendsthe token to(տ
orւ
)andinright,to(ր
orց
). Thememory swit hesfromright toleft with probabilityone when the token visits ac
,and from left to right withprobability1
2
at ea h step.4.1.
C
is winning for AdamInthe ase where Adam winstheset
C
,the onstru tion ofσ
relieson Lemma 4.3: Lemma 4.3. LetF ⊆ P(C)
be a Muller winning ondition su h thatC /
∈ F
, andA
be a 21
2
-player arena su h that Eve winseverywhere. There are subarenasA
1
. . . A
n
su h that:• i 6= j ⇒ A
i
∩ A
j
= ∅
;• ∀i, A
i
is a trap for Adam inthe subarenaA \ Attr
E
∪
i−1
j=1
A
j
;
• ∀i, χ(A
i
)
is in luded in the labelE
i
of a hild of the root ofZ
F ,C
, and Eve wins everywhere in(A
i
, F ↾ E
i
)
;• A = Attr
E
(∪
n
j=1
A
j
)
.Letthe subarenas
A
i
bethe oneswhose existen eis proved inthis lemma. We denote byσ
i
thealmost-sure strategyfor Eve inA
i
,and bya
i
theattra torstrategy for EvetoA
i
inthearenaA \ Attr(∪
i−1
j=1
A
j
)
. We identify the memorystates oftheσ
i
,sotheirunionhas thesame ardinalas thelargest of them. For a states
,ifi = min{j | s ∈ Attr
E
(∪
j
k=1
A
k
)}
, we deneσ(s, m)
by:•
ifs ∈ A
i
σ
u
(s, m) = σ
u
i
(s, m)
σ
n
(s, m) = σ
n
i
(s, m)
•
ifs ∈ Attr
E
(∪
i
k=1
A
k
) \ A
i
σ
u
(s, m) = m
σ
n
(s, m) = a
i
(s)
Byindu tion hypothesis overthe number of olours, we an assumethatthestrategies
σ
i
haver
F
i
memorystates. Thestrategyσ
usesmax{r
F
i
}
memory states. Proposition 4.4.P
σ,τ
Proof. The subarenas
A
i
are embedded traps, dened in su h a way that the token an es ape anA
i
only by going to the attra tor of a smaller one. Eve has thus a positive probabilityofrea hinganA
j
withj < i
. Thus, ifthetokenes apesoneof theA
i
innitely often, thetoken hasprobabilityone to go toanA
j
withj < i
. Byargument ofminimality, aftera nite prex,the token will stayinone of the trapsforever.The strategy
σ
i
is almost-sure from any state inA
i
. As Muller onditions are prex-independent, it follows from Proposition 4.4 thatσ
is also almost-sure from any state inA
.4.2.
C
iswinningfor Eve,and the rootofZ
F ,C
has no leaves among its hildren. Inthis ase, the onstru tion relies onthe following lemma:Lemma 4.5. Let
F ⊆ P(C)
be a Muller winning ondition su h thatC ∈ F
,A
a 21
2
-player arena oloured byC
su hthat Eve winseverywhere, andA
i
the labelof a hild of therootinZ
F ,C
. Then,Evewinseverywhere onthesubarenaA \ Attr
E
(χ
−1
(C \ A
i
))
withthe onditionF ↾ A
i
.Eve has a strategy
σ
i
that is almost-sure from ea h state inA \ Attr
E
(χ
−1
(C \ A
i
))
. In this ase, the set of memory states ofσ
isM = ∪
k
i=1
(i × M
i
)
. The next-move and memory-update fun tionsσ
n
and
σ
u
for a memory state
m = (i, m
i
)
aredened below:
•
ifs ∈ χ
−1
(C \ A
i
)
σ
u
(s, (i, m
i
)) = (i + 1, m
i+1
)
wherem
i+1
isanystate inM
i+1
ifs ∈ S
E
,σ
n
(s, (i, m
i
))
isany su essorofs
inA
•
ifs ∈ Attr
E
(χ
−1
(C \ A
i
))
σ
u
(s, (i, m
i
)) = (i, m
i
)
σ
n
(s, (i, m
i
)) = a
i
(s)
•
ifs ∈ A \ Attr
E
(χ
−1
(C \ A
i
))
σ
u
(s, (i, m
i
)) = (i, σ
u
i
(s, m
i
))
σ
n
(s, (i, m
i
)) = σ
n
i
(s, m
i
)
On eagain, we anassumethatthememory
M
i
ofthestrategyσ
i
isofsizer
F ↾A
i
. Here, however, the memory set ofσ
is thedisjoint unionof theM
i
',soσ
's needsthe sumof the{r
F ↾A
i
}
's.Proposition 4.6. Let
uc
be the event the top-level memory ofσ
is ultimately onstant. Then,P
σ,τ
s
0
(ρ ∈ Φ
F
| uc) = 1
.Proof. We all
i
thevalueofthetop-levelmemoryatthelimit. Afteraniteprex,thetoken stopsvisitingχ
−1
(C \ A
i
)
. Thus,withprobabilityone,italsostops visitingAttr
E
(χ
−1
(C \
A
i
))
. Fromthispointon,thetokenstaysinthearenaA
i
,whereEveplays withthe almost-sure strategyσ
i
. Thus,P
σ,τ
(ρ ∈ Φ
F ↾A
i
| uc) = 1
, and, asΦ
F ↾A
i
⊆ Φ
F
, Proposition 4.6 follows.Proposition 4.7. If the top-level memory takes ea h value in
1 . . . k
innitely often, then surely,∀i ∈ 1 . . . k, χ(Inf(ρ))
* A
i
.Proof. The update on the top-level memory follows a y le on
1 . . . k
, leavingi
only when thetokenvisitsχ
−1
(C \A
i
)
. Thus,inorderforthetop-levelmemoryto hange ontinuously, thetokenhasto visit ea h oftheχ
−1
(C \ A
4.3.
C
iswinningforEve,andtherootofZ
F ,C
hasatleast oneleafinits hildren. As in the previous se tion, the onstru tion relies on Lemma 4.5. In fa t, the on-stru tion for hildren whi h are not leaves, labelledA
1
, . . . , A
k
, is exa tly the same. The dieren eisthatweaddhereasinglememorystate0
thatrepresentsalltheleaves (la-belledA
−1
, . . . , A
−l
). Thememorystatesarethusupdatedmodulok +1
,andnotmodulok
. The next-move fun tion ofσ
whenthe top-level memory is0
is aneven distribution over all thesu essorsinA
of the urrent state. The memory-update fun tion hasprobability1
2
to stay into0
, and1
2
to go to(1, m
1
)
, for some memory statem
1
∈ M
1
. Thus,σ
uses memoryP
k
i=1
r
F
i
+ 1
. We prove now thatσ
is almost-sure. The stru ture of the proof is thesame asintheformer se tion, withsome extra onsiderations for thememory state0
. Proposition4.8. Letuc
be the eventthetop-level memory ofσ
isultimately onstant and dierentfrom 0. Then,P
σ,τ
s
(ρ ∈ Φ
F
| uc) = 1
.Proof. The proofis exa tlythe sameastheone of Proposition4.6.
Proposition 4.9. The event the top-level memory is ultimately onstant and equal to 0 has probability 0.
Proof. When the top-level memory is 0, the memory-update fun tion has probability
1
2
at ea h stepto swit hto 1. Proposition 4.9follows.Proposition 4.10 onsiders the ase where the top-level memory evolves ontinuously. By denition of the memory update, this an happen only if all the memory states are visitedinnitelyoften.
Proposition 4.10. Let
ec
be the event the top-level memory takes ea h value in0 . . . k
innitely often. Then,∀i ∈ −l . . . k,
P
σ,τ
s
(χ(Inf(ρ)) ⊆ C
i
| ec) = 0
.Proof. Asintheproofof Proposition 4.7,fromthefa tthatthememory isequalto ea hof the
i ∈ 1 . . . k
innitelyoften, we an dedu ethatthe token surely visitsea h oftheC \ A
i
innitely often. We only need to show that, with probability one and for anyj ∈ 1 . . . l
, thesetoflimit statesisnot in luded inA
−j
. The ZielonkaTrees ofthe onditionsF ↾ A
−j
areleaves. Thismeans that they aretrivial onditions,where allthe playsare winning for Adam. Consequently, in this ase, Lemma 4.5 guarantees thatAttr
E
(χ
−1
(C \ A
−j
))
isthe wholearena. Thedenitionofσ
inthe memory state(0)
isto playlegal movesat random. ThereisthusapositiveprobabilitythatEvewillplaya ordingtotheattra torstrategya
j
long enough to guarantee a positive probability that thetoken visitsχ
−1
(C \ A
−j
)
. To be pre ise, for anys ∈ S
,this probability is greater than(2 · |S|)
−|S|
. Thus, withprobability one,thetoken visitsea h
χ
−1
(C \ A
−j
)
innitelyoften. Proposition 4.10follows.Theinitial ase, where theZielonkatreeis redu edto aleaf, istrivial: thewinnerdoes not depend ontheplay. Thus, Theorem 4.2followsfrom Se tions4.1, 4.2, and4.3.
5. Lower Bound
Theorem 5.1 ([Maj03℄). For any set of olours
C
, there is a 2-player Muller gameG
C
=
(A
C
, F
C
)
su h that Eve has an almostsure, but none of heralmost-sure strategies have less than|C|
2
!
memory states.However, this is a general lower bound on all Muller onditions, while we aim to nd spe i lowerboundsforea h ondition. We prove herethatthereisalowerboundforea h Muller onditionthat mat hes the upperboundof Theorem 4.2:
Theorem 5.2. Let
F
be a Muller ondition onC
. There is a 2-player arenaA
F
overC
su h that Eve has a sure strategy, but none of her almost-sure strategies have less thanr
F
memory states.Asthe onstru tionoftheupperboundwasbasedontheZielonkatree,thelowerbound isbased onthe Zielonka DAG:
Denition 5.3. The ZielonkaDAG
D
F ,C
ofawinning onditionF ⊆ P(C)
isderived fromZ
F ,C
bymergingthe nodeswhi hshare thesame label.5.1. Cropped DAGs
Therelation between
r
F
and theshapeofD
F ,C
isasymmetri al: itdependsdire tlyon thenumber of hildrenof Eve's nodes, andnot at all onthe numberof hildren of Adam's nodes. Thenotion of ropped DAG isthe nextlogi al step: a sub-DAGwhere Eve's nodes keep all their hildren, while ea h nodeof Adamkeepsonly one hild:Denition 5.4. ADAG
E
isa ropped DAG ofa ZielonkaDAGD
F ,C
ifand onlyif•
The nodes ofE
arenodesofD
F ,C
,with thesame owner andlabel.•
There isonly one node withoutprede essor inE
, whi h we all the root ofE
. It is the root ofD
F ,C
,ifitbelongs to Eve;otherwise, itis oneof its hildren.•
The hildren ofa node of EveinE
areexa tlyits hildren inD
F ,C
.•
A node of Adam has exa tly one hild inE
, hosen among his hildren inD
F ,C
, provided there isone. If ithasno hildreninD
F ,C
,ithasno hildren inE
.CroppedDAGresembleZielonka DAGs: thenodesbelong to eitherEve or Adam, and they arelabelled by setsof states. We an thus ompute thenumber
r
E
ofa roppedDAGE
inanaturalway. Infa t,this numberhasamoreintuitivemeaninginthe aseof ropped DAGs: iftheleavesbelongtoEve,itisthenumberof bran hes;ifAdamownstheleaves,it isthenumberofbran heswiththeleafremoved. Furthermore,thereisadire tlinkbetween the ropped DAGsof aZielonka DAGD
F ,C
andthe numberr
F
:Proposition 5.5. Let
F
be a Muller ondition onC
, andD
F ,C
be its Zielonka DAG.Then there isa ropped DAGE
∗
su h that
r
E
∗
= r
F
.5.2. From ropped DAGs to arenas
From any ropped DAG
E
ofD
F ,C
, we dene an arenaA
E
whi h follows roughly the stru ture ofE
: the token starts from the root, goes towards the leaves, and then restarts fromtheroot. Inhernodes,Eve an hoosetowhi h hildshewantstogo. Adam's hoi es, ontheotherhand, onsistsineitherstoppingthe urrenttraversalorallowingittopro eed.•
inPick
∗
(C)
,Adam an visit anysubset of oloursin
C
;•
inPick(D)
,he mustvisit exa tlyone olour inD
.Bothare representedinFigure2,and theyaretheonly o asionswhere oloursarevisited in
A
E
: all the otherstates are olourless.c
1
c
i
c
k
· · ·
· · ·
C = {c
1
. . . c
k
}
(a)Pick
∗
(C)
d
1
· · ·
d
i
· · ·
d
k
D = {d
1
. . . d
k
}
(b)Pick(D)
Figure2:Pick
∗
(C)
andPick(D)
Eve'sstatesinthearena
A
E
areinbije tionwithhernodesinE
. Adam'snodes,onthe other hand, are in bije tion with the pairs parent- hild ofE
, where the parent belongs to Eve and the hild to Adam.In the state orresponding to the node
n
, Eve an send the token to any state of the formn − c
. In states orresponding to leaves, Eve has no de ision to take, and Adam an visit any olours inthe labelof the leaf(Pick
∗
pro edure). The token isthen sent ba k to theroot.
Adam'smoves donot involve the hoi e of a hild: by Denition5.4, Adam'snodesin
E
have but one hild. Instead, he an either stop the urrent traversal, or, if the urrent node isnot a leaf, allow itto pro eed to its only hild. If he hooses to stop, Adamhas to visit some oloured states before the token is sent ba kto the root. The available hoi es depend on the labels of both the urrent and the former nodes whi h is why there are asmany opies of Adam's nodes inA
E
as they have parents inE
. If theparent islabelled byE
,andthe urrent node byA
,thetokengoesthroughPick
∗
(E)
and
Pick(E \ A)
. Adam anthus hooseanynumberof oloursinE
,aslongashe hoosesat leastoneoutsideofA
.E
A
E
′
E
E
− A
E
′
Pick
∗
(E)
Pick(E \ A)
root(a)Edge
E
-A
whenA
isanodeE
A
Pick
∗
(E)
Pick(E \ A)
root
(b)Edge
E
-A
whenA
isaleafFigure 3: Adam'sstates in
A
E
.5.3. Winning strategy and bran h strategies
b = E
1
A
1
. . . E
ℓ
(A
ℓ
)
, Eve's moves follow the bran hb
: inE
i
, she goes toE
i
− A
i
. When Adamstops the traversalat thei
th step, Eveupdates hermemory asfollows:•
IfE
i
haszero orone hild inE
,thememoryis un hanged;•
otherwise, thenewmemorybran hhasE
1
A
1
. . . E
i
A
asaprex,whereA
isthenext hild ofE
i
,or therst one ifA
i
wasthe last.Proposition 5.6. The strategy
ς
is surely winning for Eve inthe game(A
E
, F)
.Proof. Let
ρ
be a play onsistent withς
. We denote byi
the smallest integer su h that traversals stopsinnitelyoftenatthei
th step. After aniteprex,therst2i − 1
nodesin the memory bran h are onstant, and we denote them byE
1
A
1
E
2
. . . E
i
. From this point on, the oloursvisited belong toE
i
. Furthermore, ea h time a traversal stops at stepi
, a stateisvisitedoutsideofthe urrentA
i
,whi h hangesafterwards tothenext,ina ir ular way. It follows thatInf(ρ) ⊆ E
i
, and, for any hildA
ofE
i
inE
,Inf(ρ)
* A
. Thusρ
is winningfor Eve. Proposition 5.6follows.Obviously, Adam has no winning strategy in
A
E
. However, we des ribe the lass of bran h strategies, whose point is to punish any attempt of Eve to win with less thanr
F
memory states. There is one su h strategyτ
b
for ea h bran hb
inE
(when e the name), and theprin iple isthatτ
b
stops thetraversalassoon asEve deviates fromb
:Denition 5.7. The bran h strategy
τ
b
for Adam inA
E
, orresponding to the bran hb = E
1
A
1
E
2
. . . E
ℓ
(A
ℓ
)
inE
,is a positionalstrategy whose moves aredes ribed below.•
In astateE − A
su hthat∃i, E = E
i
∧ A 6= A
i
: stop thetraversaland visitA
i
;•
ina stateE − A
su hthat∃i, E = E
i
∧ A = A
i
: sendthetoken toE
i+1
;•
inthe stateE
ℓ
− A
ℓ
,or the leafE
ℓ
: visit the oloursofE
ℓ
.Nomoveisgivenforastate
E −A
su hthat∀i, E 6= E
i
,asthesestatesarenotrea hable fromtheroot whenAdamplaysτ
b
. Noti ealsothatwhenAdam hooses tostopatraversal ina stateE
i
− A
,he an visit exa tlythe olours ofA
i
: asA
andA
i
aremaximal subsets ofE
i
,thereisat leastone state inA
i
\ A
thathe an pi kinthePick(E
i
\ A)
area.5.4. Winning against bran h strategies
The key ideaof the proof of Theorem 5.2 isthat iftwo bran hes
b
andb
′
of
E
are too dierent, Eve needsdierent memory statesto win againstτ
b
andτ
b
′
.Proposition 5.8. Let
σ = (M, σ
n
, σ
u
)
be an almost-sure strategy forEve in(A
E
, F)
. Thenσ
has memory atleastr
E
.Proof. Let
b = E
1
A
1
. . . E
ℓ
(A
ℓ
)
beabran hofE
andτ
b
bethe orrespondingbran hstrategy for Adam. By denition ofτ
b
,the set of oloursvisited ina traversal onsistent withτ
b
is one oftheA
i
's, orE
ℓ
ifand onlyifEve playsalongb
. Asσ
isalmost-sure,there mustbe a memorystatem
su hthatEvehasapositiveprobabilitytoplayalongb
. Itisalsone essary to ensure thatnone oftheA
i
'sis visitedinnitelyoften, withthepossible ex eption ofA
ℓ
. So, if Eve has a positive to play along a bran hb
′
when she is in the memory state
m
,E
1
A
1
. . . E
ℓ
must be a prex ofb
′
. It follows that a single memory state an be suitable against two strategies
τ
b
andτ
b
′
withb = E
1
A
1
. . . E
ℓ
(A
ℓ
)
andb
′
= E
′
1
A
′
1
. . . E
ℓ
′
′
(A
′
ℓ
′
)
only ifℓ = ℓ
′
and∀i ≤ ℓ, E
i
= E
′
By Proposition 5.5, there is a ropped DAG
E
ofD
F ,C
su h thatr
E
= r
F
. So, in general, Eve needs randomised strategies with memoryr
F
in order to win games whose winning ondition isF
. This ompletes theproofof Theorem5.2.6. Con lusion
Wehaveprovidedbetterandtightboundsforthememoryneededto denealmostsure winningrandomised strategies. Thisallowsus to hara terisethe lassofMuller onditions whi h admit randomisedmemoryless strategies:
Corollary6.1. Eveadmitsrandomised memorylessalmost-surestrategies foraMuller on-dition
F
if andonly ifall her nodes inZ
F ,C
have either one hild, or onlyleave hildren.This yields a NPalgorithm for the winnerproblem of su h games, assolving 1
1
2
-player MullergamesisPtime[CdAH04℄. Another onsequen eofourresultisthatforea hMuller ondition, at leastone ofthe players annotimprove its memory throughrandomisation: Corollary 6.2. LetF
be a Muller ondition. If∅ ∈ F
, Eve needs as mu h memory for randomised strategy as for pure strategies. Otherwise, Adam does.Our proof of lower bound also improves on the size of thewitness arena: itis roughly equivalentto thesizeoftheZielonkaDAG,insteadofthesizeoftheZielonkatree. Whether these bounds still hold for arenas of size polynomial in the number of olours remains an openquestion, ex eptforspe ial ases like Streett games[Hor07 ℄.
Referen es
[BL69℄ J.Ri hardBü hiandLawren eH.Landweber.Solvingsequential onditionsbynite-state strate-gies.TAMS,138:295311, 1969.
[CdAH04℄ KrishnenduChatterjee,Lu adeAlfaro,andThomasA.Henzinger.Tradingmemoryfor random-ness.InQEST,pages206217,2004.
[Cha07a℄ KrishnenduChatterjee.OptimalStrategySynthesisinSto hasti MullerGames.InFOSSACS, pages138152,2007.
[Cha07b℄ KrishnenduChatterjee.Sto hasti MüllerGamesarePSPACE-Complete.InFSTTCS,pages436 448,2007.
[deA97℄ Lu adeAlfaro.FormalVeri ationofProbabilisti Systems.PhDthesis,Stanford,1997. [DJW97℄ StefanDziembowski,Mar inJurdzinski,andIgorWalukiewi z.Howmu hmemoryisneededto
wininnitegames? InLICS,pages99110,1997.
[GH82℄ YuriGurevi handLeoHarrington.Trees,automata,andgames.InSTOC,pages6065,1982. [Hor07℄ FlorianHorn.Di ingontheStreett.IPL,104(1):19,2007.
[Maj03℄ RupakMajumdar.Symboli Algorithms forVeri ationandControl.PhDthesis,Berkeley,2003 [MP92℄ ZoharMannaandAmirPnueli.The temporalLogi ofRea tiveandCon urrentSystems:
Spe i- ation.Springer,1992.
[PR89℄ AmirPnueli andRoniRosner.Onthesynthesisofarea tivemodule.InPOPL,pages179190, 1989.
[Tho95℄ WolfgangThomas.Onthesynthesisofstrategiesininnitegames.InSTACS,pages113,1995. [Zie98℄ WieslawZielonka.Innitegamesonnitely olouredgraphswith appli ations toautomataon
innitetrees.TCS,200(1-2):135183, 1998.