Abstract | ||
---|---|---|
We introduce a new embarrassingly parallel parameter learning algorithm for Markov random fields with untied parameters which is efficient for a large class of practical models. Our algorithm parallelizes naturally over cliques and, for graphs of bounded degree, its complexity is linear in the number of cliques. Unlike its competitors, our algorithm is fully parallel and for log-linear models it is also data efficient, requiring only the local sufficient statistics of the data to estimate parameters. |
Year | Venue | Field |
---|---|---|
2014 | ICML | Graph,Parallel learning,Random field,Computer science,Embarrassingly parallel,Markov chain,Artificial intelligence,Log-linear model,Sufficient statistic,Machine learning,Bounded function |
DocType | Citations | PageRank |
Conference | 7 | 0.58 |
References | Authors | |
29 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yariv Dror Mizrahi | 1 | 8 | 0.93 |
Misha Denil | 2 | 397 | 26.18 |
Nando De Freitas | 3 | 3284 | 273.68 |