Non-Count Symmetries in Boolean & Multi-Valued Prob. Graphical Models
نویسندگان
چکیده
Lifted inference algorithms commonly exploit symmetries in a probabilistic graphical model (PGM) for efficient inference. However, existing algorithms for Boolean-valued domains can identify only those pairs of states as symmetric, in which the number of ones and zeros match exactly (count symmetries). Moreover, algorithms for lifted inference in multi-valued domains also compute a multi-valued extension of count symmetries only. These algorithms miss many symmetries in a domain. In this paper, we present first algorithms to compute non-count symmetries in both Booleanvalued and multi-valued domains. Our methods can also find symmetries between multi-valued variables that have different domain cardinalities. The key insight in the algorithms is that they change the unit of symmetry computation from a variable to a variable-value (VV) pair. Our experiments find that exploiting these symmetries in MCMC can obtain substantial computational gains over existing algorithms.
منابع مشابه
χChek: A Model Checker for Multi-Valued Reasoning
This paper describes our multi-valued symbolic modelchecker χChek. χChek is a generalization of an existing symbolic model-checking algorithm for a multi-valued extension of the temporal logic CTL. Multi-valued modelchecking supports reasoning with values other than just TRUE and FALSE. Multi-valued logics are useful in software engineering because they support explicit modeling of uncertainty,...
متن کاملA High-Level Petri Net Framework for Multi-Valued Genetic Regulatory Networks
To understand the function of genetic regulatory networks in the development of cellular systems, we must not only realise the individual network entities, but also the manner by which they interact. Multi-valued networks are a promising qualitative approach for modelling such genetic regulatory networks, providing an interesting compromise between the simplicity of Boolean models and more deta...
متن کامل\chiChek: A Model Checker for Multi-Valued Reasoning
This paper describes our multi-valued symbolic modelchecker χChek. χChek is a generalization of an existing symbolic model-checking algorithm for a multi-valued extension of the temporal logic CTL. Multi-valued modelchecking supports reasoning with values other than just TRUE and FALSE. Multi-valued logics are useful in software engineering because they support explicit modeling of uncertainty,...
متن کاملModel Checking for Multi - Valued CTL
A multi-valued version of CTL (mv-CTL), where both the propositions and the accessibility relation are multi-valued taking values in a nite quasi-Boolean algebra, is deened. A translation from mv-CTL model checking to CTL model checking is investigated. First, the case where the elements of quasi-Boolean algebras are totally ordered is considered. Secondly, it is shown how to design a translati...
متن کاملMultiplicative Factorization of Multi-Valued NIN-AND Tree Models
A multi-valued Non-Impeding Noisy-AND (NIN-AND) tree model has the linear complexity and is more expressive than common Causal Independence Models (CIMs). We formulate a Multiplicative Factorization (MF) for multi-valued NIN-AND Tree (NAT) models. In comparison with the MF for binary NAT models (of a undirected tree structure), the proposed MF is a hybrid and multiply connected graphical model....
متن کامل