0 # initial state
(0,0,'c',ConstantRewardFunction(0))
(0,1,'d',ConstantRewardFunction(0))
(0,3,'a',ConstantRewardFunction(0))
(0,4,'e',ConstantRewardFunction(1))
(1,2,'b',ConstantRewardFunction(0))
(1,3,'c|d',ConstantRewardFunction(0))
(1,4,'e',ConstantRewardFunction(1))
(2,0,'c',ConstantRewardFunction(0))
(2,1,'d',ConstantRewardFunction(0))
(2,3,'b',ConstantRewardFunction(0))
(2,4,'e',ConstantRewardFunction(1))
(4,4,'True',ConstantRewardFunction(0))