Advances in Bayesian Network Learning using Integer Programming
arXiv:1309.6825
Abstract
We consider the problem of learning Bayesian networks (BNs) from complete discrete data. This problem of discrete optimisation is formulated as an integer program (IP). We describe the various steps we have taken to allow efficient solving of this IP. These are (i) efficient search for cutting planes, (ii) a fast greedy algorithm to find high-scoring (perhaps not optimal) BNs and (iii) tightening the linear relaxation of the IP. After relating this BN learning problem to set covering and the multidimensional 0-1 knapsack problem, we present our empirical results. These show improvements, sometimes dramatic, over earlier results.
Appears in Proceedings of the Twenty-Ninth Conference on Uncertainty in Artificial Intelligence (UAI2013)
References in corpus (4)
Cited by in corpus (13)
- Optimization Problems for Machine Learning: A Survey
- A hybrid algorithm for Bayesian network structure learning with application to multi-label learning
- Searching Multiregression Dynamic Models of Resting-State fMRI Networks Using Integer Programming
- Marginal Pseudo-Likelihood Learning of Markov Network structures
- Advances in Learning Bayesian Networks of Bounded Treewidth
- Learning Chordal Markov Networks by Constraint Satisfaction
- Estimating causal structure using conditional DAG models
- Consistent Second-Order Conic Integer Programming for Learning Bayesian Networks
- Structure Learning in Bayesian Networks of Moderate Size by Efficient Sampling
- Towards a Multi-Subject Analysis of Neural Connectivity
- A Score-and-Search Approach to Learning Bayesian Networks with Noisy-OR Relations
- Scalable Bayesian Network Structure Learning with Splines
- Learning All Credible Bayesian Network Structures for Model Averaging