FSPN: A New Class of Probabilistic Graphical Model
arXiv:2011.09020
Abstract
We introduce factorize sum split product networks (FSPNs), a new class of probabilistic graphical models (PGMs). FSPNs are designed to overcome the drawbacks of existing PGMs in terms of estimation accuracy and inference efficiency. Specifically, Bayesian networks (BNs) have low inference speed and performance of tree structured sum product networks(SPNs) significantly degrades in presence of highly correlated variables. FSPNs absorb their advantages by adaptively modeling the joint distribution of variables according to their dependence degree, so that one can simultaneously attain the two desirable goals: high estimation accuracy and fast inference speed. We present efficient probability inference and structure learning algorithms for FSPNs, along with a theoretical analysis and extensive evaluation evidence. Our experimental results on synthetic and benchmark datasets indicate the superiority of FSPN over other PGMs.
arXiv comment16 pages
References in corpus (9)
- Loopy Belief Propagation for Approximate Inference: An Empirical Study
- Sum-Product Networks: A New Deep Architecture
- DAGs with NO TEARS: Continuous Optimization for Structure Learning
- DAG-GNN: DAG Structure Learning with Graph Neural Networks
- The Randomized Dependence Coefficient
- On the Expressive Efficiency of Sum Product Networks
- Conditional Sum-Product Networks: Imposing Structure on Deep Probabilistic Architectures
- Tractable Querying and Learning in Hybrid Domains via Sum-Product Networks
- Sum-product networks: A survey