FSPN: A New Class of Probabilistic Graphical Model
This work addresses the problem of improving estimation accuracy and inference efficiency for users of probabilistic graphical models, presenting an incremental advancement in the field.
This paper introduces Factorize Sum Split Product Networks (FSPNs), a new class of probabilistic graphical models. FSPNs aim to improve upon existing PGMs by adaptively modeling variable dependencies to achieve both high estimation accuracy and fast inference speed, outperforming other PGMs on synthetic and benchmark datasets.
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.