Communication-Avoiding Optimization Methods for Massive-Scale Graphical Model Structure Learning.
Penporn KoanantakoolAlnur AliAriful AzadAydin BuluçDmitriy MorozovSang-Yun OhLeonid OlikerKatherine A. YelickPublished in: CoRR (2017)
Keyphrases
- structure learning
- optimization methods
- graphical models
- massive scale
- belief propagation
- probabilistic model
- optimization method
- bayesian networks
- conditional independence
- optimization problems
- random variables
- probabilistic inference
- approximate inference
- markov logic networks
- markov networks
- simulated annealing
- probabilistic graphical models
- parameter learning
- conditional random fields
- exact inference
- low latency
- dynamic bayesian networks
- undirected graphical models
- belief networks
- factor graphs
- genetic algorithm
- text mining
- upper bound