# This is a Dec-POMDP (.dpomdp) file for the Dec-Tiger problem. # For more detailed documentation, see example.dpomdp # Allright, here we go! # #The agents. #---------- #Either 1) the number of agents: # agents: %d #or 2) a list of agent identifiers, e.g.: # agents: agent1_name, name-of-agent2, ... agents: 2 # discount: %f discount: 1 #.0 # values: [ reward, cost ] values: reward # states: [ %d, ] states: tiger-left tiger-right # #Examples of this are: # start: 0.3 0.1 0.0 0.2 0.5 # start: first-state # start: 5 # start: uniform # start include: first-state third state # start include: 1 3 # start exclude: fifth-state seventh-state start: uniform # #The actions declarations #------------------------ #the (number/list of) actions for each of the agents on a separate line # actions: # [ %d, ] # [ %d, ] # ... # [ %d, ] actions: listen open-left open-right listen open-left open-right #the (number/list of) observations for each of the agents on a separate line # observations: # [ %d, ] # [ %d, ] # ... # [ %d, ] observations: hear-left hear-right hear-left hear-right #Transition probabilities # T: : : : %f #or # T: : : # %f %f ... %f P(s_1'|ja,s) ... P(s_k'|ja,s) #or # T: : this is a |S| x |S| matrix # %f %f ... %f P(s_1'|ja,s_1) ... P(s_k'|ja,s_1) # %f %f ... %f ... # ... ... # %f %f ... %f P(s_1'|ja,s_k) ... P(s_k'|ja,s_k) #or # T: # [ identity, uniform ] T: * : uniform #T:open-right open-right : #uniform T: listen listen : identity #Observation probabilities # O: : : : %f #or # O: : : # %f %f ... %f P(jo_1|ja,s') ... P(jo_x|ja,s') #or # O: : - a |S|x|JO| matrix # %f %f ... %f P(jo_1|ja,s_1') ... P(jo_x|ja,s_1') # %f %f ... %f ... # ... ... # %f %f ... %f P(jo_1|ja,s_k') ... P(jo_x|ja,s_k') O: * : uniform O: listen listen : tiger-left : hear-left hear-left : 0.7225 O: listen listen : tiger-left : hear-left hear-right : 0.1275 O: listen listen : tiger-left : hear-right hear-left : 0.1275 O: listen listen : tiger-left : hear-right hear-right : 0.0225 O: listen listen : tiger-right : hear-right hear-right : 0.7225 O: listen listen : tiger-right : hear-left hear-right : 0.1275 O: listen listen : tiger-right : hear-right hear-left : 0.1275 O: listen listen : tiger-right : hear-left hear-left : 0.0225 #The rewards #or # R: : : : # %f %f ... %f #or # R: : : # %f %f ... %f # %f %f ... %f # ... # %f %f ... %f # #Typical problems only use R(s,ja) which is specified by: # R: : : * : * : %f R: listen listen: * : * : * : -2 R: open-left open-left : tiger-left : * : * : -50 R: open-right open-right : tiger-right : * : * : -50 R: open-left open-left : tiger-right : * : * : +20 R: open-right open-right : tiger-left : * : * : 20 R: open-left open-right: tiger-left : * : * : -100 R: open-left open-right: tiger-right : * : * : -100 R: open-right open-left: tiger-left : * : * : -100 R: open-right open-left: tiger-right : * : * : -100 R: open-left listen: tiger-left : * : * : -101 R: listen open-right: tiger-right : * : * : -101 R: listen open-left: tiger-left : * : * : -101 R: open-right listen: tiger-right : * : * : -101 R: listen open-right: tiger-left : * : * : 9 R: listen open-left: tiger-right : * : * : 9 R: open-right listen: tiger-left : * : * : 9 R: open-left listen: tiger-right : * : * : 9