teams- Graphical modeling tools for sequential teamsSource codeContentsIndex
Determine conditional independence
Structural results for teams
This module implments an automated algortihm to simplify sequential teams. The simplification is based on conditional independences. Conditional independence is checked using the Bayes Ball algorithm
module Data.Teams.Graph
observations :: Team -> [Node] -> [Node] -> [Node]
irrelevant :: Team -> [Node] -> [Node] -> [Node]
determined :: Team -> [Node] -> [Node]
effective :: Team -> [Node] -> [Node] -> [Node]
simplifyAt :: Team -> Node -> Team
simplifyOnce :: Team -> Team
simplify :: Team -> Team
module Data.Teams.Graph
Determine conditional independence
observations :: Team -> [Node] -> [Node] -> [Node]Source
Requisite observations The observation nodes are thouse nodes in the condition that are marked as visited
irrelevant :: Team -> [Node] -> [Node] -> [Node]Source
Irrelevant nodes The nodes that have not been visited from their parents are irrelevant
determined :: Team -> [Node] -> [Node]Source
Functionally determined nodes Nodes that are irrelevant when we want to know about all variable nodes
effective :: Team -> [Node] -> [Node] -> [Node]Source
Effectively observed nodes All the ancestors of the reward nodes that are functionally determined by conditioned nodes.
Structural results for teams
simplifyAt :: Team -> Node -> TeamSource
The graph restructuring algorithm of the paper.
simplifyOnce :: Team -> TeamSource
Simplify all nodes of the graph once
simplify :: Team -> TeamSource
The graph simplification aglorithm of the paper I believe that this algorithm will always converge. So, I do not stop the loop after a finite number of iterations. If you find an example that does not converge, please let me know.
Produced by Haddock version 2.4.2