Optimized Realization of Bayesian Networks in Reduced Normal Form using Latent Variable Model
This work addresses efficiency bottlenecks for practitioners using Bayesian networks in FGrn, though it appears incremental as it builds on existing methods with optimizations.
The paper tackles the high computational and memory costs of Bayesian networks in Factor Graph Reduced Normal Form (FGrn), which limit their practical use, by proposing various cost-reduction solutions and an efficient C++ library. Results show the library is quite efficient compared to standard sum-product and Maximum Likelihood algorithms, with an online learning algorithm achieving very similar results to batch learning in unsupervised contexts.
Bayesian networks in their Factor Graph Reduced Normal Form (FGrn) are a powerful paradigm for implementing inference graphs. Unfortunately, the computational and memory costs of these networks may be considerable, even for relatively small networks, and this is one of the main reasons why these structures have often been underused in practice. In this work, through a detailed algorithmic and structural analysis, various solutions for cost reduction are proposed. An online version of the classic batch learning algorithm is also analyzed, showing very similar results (in an unsupervised context); which is essential even if multilevel structures are to be built. The solutions proposed, together with the possible online learning algorithm, are included in a C++ library that is quite efficient, especially if compared to the direct use of the well-known sum-product and Maximum Likelihood (ML) algorithms. The results are discussed with particular reference to a Latent Variable Model (LVM) structure.