Title
Randomness for Free
Abstract
We consider two-player zero-sum games on graphs. These games can be classified on the basis of the information of the players an d on the mode of interaction between them. On the basis of information the classification is as fol- lows: (a) partial-observation (both players have partial v iew of the game); (b) one-sided complete-observation (one player has complete observation); and (c) complete-observation (both players have complete view of the game). On the ba- sis of mode of interaction we have the following classificati on: (a) concurrent (players interact simultaneously); and (b) turn-based (pl ayers interact in turn). The two sources of randomness in these games are randomness in transition func- tion and randomness in strategies. In general, randomized strategies are more powerful than deterministic strategies, and randomness in transitions gives more general classes of games. We present a complete characterization for the classes of games where randomness is not helpful in: (a) the transition function (proba- bilistic transition can be simulated by deterministic tran sition); and (b) strategies (pure strategies are as powerful as randomized strategies) . As consequence of our characterization we obtain new undecidability results for these games.
Year
DOI
Venue
2010
10.1007/978-3-642-15155-2_23
Computing Research Repository
Keywords
DocType
Volume
deterministic strategy,players interact,transition function,following classification,complete view,probabilistic transition,complete observation,complete characterization,deterministic transition,randomized strategy
Conference
abs/1006.0
Issue
ISSN
ISBN
C
0302-9743
3-642-15154-X
Citations 
PageRank 
References 
9
0.57
12
Authors
4
Name
Order
Citations
PageRank
Krishnendu Chatterjee12179162.09
Laurent Doyen297955.96
Hugo Gimbert324921.31
Thomas A. Henzinger4148271317.51