SGA: A Robust Algorithm for Partial Recovery of Tree-Structured Graphical Models with Noisy Samples.
Anshoo TandonAldric H. J. HanVincent Y. F. TanPublished in: ICML (2021)
Keyphrases
- graphical models
- probabilistic model
- graph structure
- map inference
- parameter estimation
- directed acyclic
- loopy belief propagation
- belief propagation
- conditional random fields
- tree structure
- expectation maximization
- np hard
- search space
- relational databases
- bayesian networks
- linear programming
- upper bound
- learning algorithm
- dynamic programming
- probabilistic inference
- probabilistic graphical models
- energy function
- bayesian framework
- junction tree
- search algorithm
- optimal solution