Using Markov Blankets for Causal Structure Learning
The Journal of Machine Learning Research
Learning graphical model structure using L1-regularization paths
AAAI'07 Proceedings of the 22nd national conference on Artificial intelligence - Volume 2
Learning bayesian network structure from massive datasets: the «sparse candidate« algorithm
UAI'99 Proceedings of the Fifteenth conference on Uncertainty in artificial intelligence
Hi-index | 0.00 |
A new algorithm, PCB (Partial Correlation-Based) algorithm, is presented for Bayesian network structure learning. The algorithm combines ideas from local learning with partial correlation techniques in an effective way. It reconstructs the skeleton of a Bayesian network based on partial correlation and then performs greedy hill-climbing search to orient the edges. Specifically, we make three contributions. Firstly, we give the proof that in a SEM (simultaneous equation model) with uncorrelated errors, when datasets are generated by SEM no matter what distribution disturbances subject to, we can use partial correlation as the criterion of CI test. Second, we have done a series of experiments to find the best threshold value of partial correlation. Finally, we show how partial relation can be used in Bayesian network structure learning under SEM. The effectiveness of the method is compared with current state of the art methods on 8 networks. Simulation shows that PCB algorithm outperforms existing algorithms in both accuracy and run time.