Second-Order Learning and Inference using Incomplete Data for Uncertain Bayesian Networks: A Two Node Example
| Autor: | Federico Cerutti, Lance M. Kaplan, Murat Sensoy, Kumar Vijay Mishra |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Random field
Training set Markov chain Computer science Posterior probability Probabilistic logic Inference Markov process Bayesian network 02 engineering and technology 010501 environmental sciences Probabilistic inference Covariance computer.software_genre 01 natural sciences symbols.namesake ComputingMethodologies_PATTERNRECOGNITION 0202 electrical engineering electronic engineering information engineering symbols 020201 artificial intelligence & image processing Data mining Fisher information computer 0105 earth and related environmental sciences |
| Zdroj: | FUSION |
| Popis: | Efficient second-order probabilistic inference in uncertain Bayesian networks was recently introduced. However, such second -order inference methods presume training over complete training data. While the expectation-maximization framework is well-established for learning Bayesian network parameters for incomplete training data, the framework does not determine the covariance of the parameters. This paper introduces two methods to compute the covariances for the parameters of Bayesian networks or Markov random fields due to incomplete data for two-node networks. The first method computes the covariances directly from the posterior distribution of parameters, and the second method more efficiently estimates the covariances from the Fisher information matrix. Finally, the implications and effectiveness of these covariances is theoretically and empirically evaluated. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|