Algorithms in Probabilistic Graphical Models – Principles and Techniques

• Algorithm 3.1 – Algorithm for finding nodes reachable from X given Z via active trails
• Algorithm 3.2 – Procedure to build a minimal I-map given an ordering
• Algorithm 3.3 – Recovering the undirected skeleton for a distribution P that has a P-map
• Algorithm 3.4 – Marking immoralities in the construction of a perfect map
• Algorithm 3.5 – Finding the class PDAG characterizing the P-map of a distribution P
• Algorithm 5.1 – Computing d-separation in the presence of deterministic CPDs
• Algorithm 5.2 – Computing d-separation in the presence of context-specific CPDs
• Algorithm 9.1 – Sum-product variable elimination algorithm
• Algorithm 9.2 – Using Sum-Product-VE for computing conditional probabilities
• Algorithm 9.3 – Maximum cardinality search for constructing an elimination ordering
• Algorithm 9.4 – Greedy search for constructing an elimination ordering
• Algorithm 9.5 – Conditioning algorithm
• Algorithm 9.6 – Rule splitting algorithm
• Algorithm 9.7 – Sum-product variable elimination for sets of rules
• Algorithm 10.1 – Upward pass of variable elimination in clique tree
• Algorithm 10.2 – Calibration using sum-product message passing in a clique tree
• Algorithm 10.3 – Calibration using belief propagation in clique tree
• Algorithm 10.4 – Out-of-clique inference in clique tree
• Algorithm 11.1 – Calibration using sum-product belief propagation in a cluster graph
• Algorithm 11.2 – Convergent message passing for Bethe cluster graph with convex counting numbers
• Algorithm 11.3 -Algorithm to construct a saturated region graph
• Algorithm 11.4 – Projecting a factor set to produce a set of marginals over a given set of scopes
• Algorithm 11.5 – Modified version of BU-Message that incorporates message projection
• Algorithm 11.6 – Message passing step in the expectation propagation algorithm
• Algorithm 11.7 – The Mean-Field approximation algorithm
• Algorithm 12.1 – Forward Sampling in a Bayesian network
• Algorithm 12.2 – Likelihood-weighted particle generation
• Algorithm 12.3 – Likelihood weighting with a data-dependent stopping rule
• Algorithm 12.4 – Generating a Gibbs chain trajectory
• Algorithm 12.5 – Generating a Markov chain trajectory
• Algorithm 13.1 – Variable elimination algorithm for MAP
• Algorithm 13.2 – Max-product message computation for MAP
• Algorithm 13.3 – Calibration using max-product BP in a Bethe-structured cluster graph
• Algorithm 13.4 – Graph-cut algorithm for MAP in pairwise binary MRFs with submodular potentials
• Algorithm 13.5 – Alpha-expansion algorithm
• Algorithm 13.6 – Efficient min-sum message passing for untruncated 1-norm energies
• Algorithm 14.1 – Expectation propagation message passing for CLG networks
• Algorithm 15.1 – Filtering in a DBN using a template clique tree
• Algorithm 15.2 – Likelihood-weighted particle generation for a 2-TBN
• Algorithm 15.3 – Likelihood weighting for filtering in DBNs
• Algorithm 15.4 – Particle filtering for DBNs
• Algorithm 18.1 – Data perturbation search
• Algorithm 19.1 – Computing the gradient in a network with table-CPDs
• Algorithm 19.2 – Expectation-maximization algorithm for BN with table-CPDs
• Algorithm 19.3 – The structural EM algorithm for structure learning
• Algorithm 19.4 – The incremental EM algorithm for network with table-CPDs
• Algorithm 19.5 – Proposal distribution for collapsed Metropolis-Hastings over data completions
• Algorithm 19.6 – Proposal distribution over partitions in the Dirichlet process priof
• Algorithm 20.1 – Greedy score-based structure search algorithm for log-linear models
• Algorithm 23.1 – Finding the MEU strategy in a decision tree
• Algorithm 23.2 – Generalized variable elimination for joint factors in influence diagrams
• Algorithm 23.3 – Iterated optimization for influence diagrams with acyclic relevance graphs
• Algorithm A.1 – Topological sort of a graph
• Algorithm A.2 – Maximum weight spanning tree in an undirected graph
• Algorithm A.3 – Recursive algorithm for computing Fibonacci numbers
• Algorithm A.4 – Dynamic programming algorithm for computing Fibonacci numbers
• Algorithm A.5 – Greedy local search algorithm with search operators
• Algorithm A.6 – Local search with tabu list
• Algorithm A.7 – Beam search
• Algorithm A.8 – Greedy hill-climbing search with random restarts
• Algorithm A.9 – Branch and bound algorithm
• Algorithm A.10 – Simple gradient ascent algorithm
• Algorithm A.11 – Conjugate gradient ascent

Last modified: Sun Oct 31 11:44:04 PDT 2010