Abstract | ||
---|---|---|
•We propose an algorithm for learning Bayesian networks with local structure.•The method is based on a logistic parametrization with interaction terms, Lasso, and an ordering-based heuristic.•Experiments with randomly generated Bayesian networks as well as standard benchmark networks are presented.•The results demonstrate good performance, and confirm the overall benefits of local structure in Bayesian networks. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1016/j.patrec.2017.06.006 | Pattern Recognition Letters |
Keywords | Field | DocType |
Bayesian networks,Context-specific independence,Sparsity,Logistic regression,Lasso | Boolean function,ENCODE,Linear combination,Conditional probability,Pattern recognition,Correctness,Lasso (statistics),Bayesian network,Artificial intelligence,Logistic regression,Mathematics | Journal |
Volume | ISSN | Citations |
95 | 0167-8655 | 0 |
PageRank | References | Authors |
0.34 | 11 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yuan Zou | 1 | 4 | 1.67 |
Pensar, Johan | 2 | 19 | 4.76 |
Teemu Roos | 3 | 436 | 61.32 |