Safe Haskell | Safe-Infered |
---|
Tools to build influence diagrams
- type InfluenceDiagram = DirectedSG EdgeKind IDValue
- type DecisionFactor = PrivateCPT Vector DVI
- class Instantiable d v where
- data DEV
- data UV
- data DV
- data TDV s
- type IDMonad g a = NetworkMonad g EdgeKind IDValue a
- t :: a
- (~~) :: (Initializable v, DirectedGraph g, Distribution d) => IDMonad g v -> d -> IDMonad g ()
- chance :: (Bounded a, Enum a, NamedGraph g) => String -> a -> IDMonad g (TDV a)
- decisionNode :: (Bounded a, Enum a, NamedGraph g) => String -> a -> IDMonad g DEV
- utilityNode :: NamedGraph g => String -> IDMonad g UV
- proba :: (ChanceVariable c, DirectedGraph g) => c -> IDMonad g c
- decision :: (DirectedGraph g, BayesianDiscreteVariable dv) => DEV -> [dv] -> IDMonad g DEV
- utility :: (DirectedGraph g, BayesianDiscreteVariable dv) => UV -> [dv] -> IDMonad g UV
- cpt :: (DirectedGraph g, BayesianDiscreteVariable vb, ChanceVariable c) => c -> [vb] -> IDMonad g c
- d :: DEV -> PorD
- p :: ChanceVariable c => c -> PorD
- noDependencies :: [DV]
- decisionsOrder :: InfluenceDiagram -> [ChancesOrDecision]
- solveInfluenceDiagram :: InfluenceDiagram -> [DecisionFactor]
- runID :: IDMonad DirectedSG a -> (a, InfluenceDiagram)
- policyNetwork :: [DecisionFactor] -> InfluenceDiagram -> SBN CPT
- decisionToInstantiation :: DecisionFactor -> [DVISet]
- type DVISet = [DVI]
- data DVI
Type
type InfluenceDiagram = DirectedSG EdgeKind IDValueSource
Influence diagram
type DecisionFactor = PrivateCPT Vector DVISource
class Instantiable d v whereSource
(Bounded b, Enum b) => Instantiable DV b | |
Instantiable DEV Int | |
(Bounded b, Enum b) => Instantiable (TDV b) b |
Decision variable
A discrete variable
A typed discrete variable
Building
Synonym for undefined because it is clearer to use t to set the Enum bounds of a variable
:: (Initializable v, DirectedGraph g, Distribution d) | |
=> IDMonad g v | Discrete variable in the graph |
-> d | List of values |
-> IDMonad g () |
chance :: (Bounded a, Enum a, NamedGraph g) => String -> a -> IDMonad g (TDV a)Source
Create a chance node
decisionNode :: (Bounded a, Enum a, NamedGraph g) => String -> a -> IDMonad g DEVSource
Create a decision node
utilityNode :: NamedGraph g => String -> IDMonad g UVSource
Create an utility node
proba :: (ChanceVariable c, DirectedGraph g) => c -> IDMonad g cSource
Define that a chance node is a probability (not conditional) Values are ordered like FFF FFT FTF FTT TFF TFT TTF TTT and same for other enumeration keeping enumeration order
decision :: (DirectedGraph g, BayesianDiscreteVariable dv) => DEV -> [dv] -> IDMonad g DEVSource
Define a decision dependence
utility :: (DirectedGraph g, BayesianDiscreteVariable dv) => UV -> [dv] -> IDMonad g UVSource
Define a utility dependence
cpt :: (DirectedGraph g, BayesianDiscreteVariable vb, ChanceVariable c) => c -> [vb] -> IDMonad g cSource
Define that a chance node is a conditional probability and define the parent variables
noDependencies :: [DV]Source
Used to define a root decision which is not dependent on any past node
Solving
decisionsOrder :: InfluenceDiagram -> [ChancesOrDecision]Source
List of decision vertices in reverse temporal order (corresponding to elimination order)
solveInfluenceDiagram :: InfluenceDiagram -> [DecisionFactor]Source
Solve an influence diagram. A DecisionFactor is generated for each decision variable. A decision factor is containing a variable instantiation instead of a double. This instantiation is giving the decision to take for each value of the parents.
runID :: IDMonad DirectedSG a -> (a, InfluenceDiagram)Source
Run an influence monad
policyNetwork :: [DecisionFactor] -> InfluenceDiagram -> SBN CPTSource
Create a policy network from an influence diagram and its solution. A policy network is a Bayesian network where the decision nodes have been replaced with probability nodes where the probability is 1 when the configuration is corresponding to the decision and 0 otherwise.
decisionToInstantiation :: DecisionFactor -> [DVISet]Source
Convert a decision policy to a set of possible instantiations It is the only way to access to the content of a decision factor.
Testing
Discrete Variable instantiation. A variable and its value
Eq DVI | |
Show DVI | |
Show MAXCPT | |
Show DecisionFactor | |
BayesianDiscreteVariable DVI | |
IsBucketItem MAXCPT | |
MultiDimTable MAXCPT | |
MultiDimTable DecisionFactor | |
Factor MAXCPT | |
LabeledVertex DVI | |
Show (DirectedSG () MAXCPT) | |
FactorElement (Double, PossibleInstantiations) |