Name
Abstract | ||
|---|---|---|
As computer clusters become more common and the size of the problems encountered in the field of AI grows, there is an increasing demand for efficient parallel inference algorithms. We consider the problem of parallel inference on large factor graphs in the distributed memory setting of computer clusters. We develop a new efficient parallel inference algorithm, DBRSplash, which incorporates over-segmented graph partitioning, belief residual scheduling, and uniform work Splash operations. We empirically evaluate the DBRSplash algorithm on a 120 processor cluster and demonstrate linear to super-linear performance gains on large factor graph models. |
Year | Venue | Keywords |
|---|---|---|
2009 | uncertainty in artificial intelligence | efficient parallel inference algorithm,belief residual scheduling,parallel inference,increasing demand,new efficient parallel inference,large factor graph model,computer cluster,over-segmented graph partitioning,dbrsplash algorithm,large factor graph |
Field | DocType | Volume |
Factor graph,Residual,Scheduling (computing),Computer science,Inference,Distributed memory,Theoretical computer science,Artificial intelligence,Graph partition,Machine learning,Computer cluster | Conference | abs/1205.2645 |
Citations | PageRank | References |
26 | 1.82 | 12 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Joseph E. Gonzalez | 1 | 2219 | 102.68 |
| Yucheng Low | 2 | 1078 | 39.38 |
| Carlos Guestrin | 3 | 9220 | 488.92 |
| David R. O'hallaron | 4 | 1243 | 126.28 |