Fast factorisation of probabilistic potentials and its application to approximate inference in Bayesian networks
Identifiers
URI: http://hdl.handle.net/10835/4885
DOI: https://doi.org/10.1142/S0218488512500110
DOI: https://doi.org/10.1142/S0218488512500110
Share
Metadata
Show full item recordDate
2012Abstract
We present an efficient procedure for factorising probabilistic potentials represented as
probability trees. This new procedure is able to detect some regularities that cannot be
captured by existing methods. In cases where an exact decomposition is not achievable,
we propose a heuristic way to carry out approximate factorisations guided by a parameter
called factorisation degree, which is fast to compute. We show how this parameter can be
used to control the tradeoff between complexity and accuracy in approximate inference
algorithms for Bayesian networks.
Palabra/s clave
Bayesian netwoorks
Probability trees
Factorisation
Probabilistic inference