Approximating Probability Density Functions with Mixtures of Truncated Exponentials
Barry R. Cobb, Prakash Pundalik Shenoy, Rafael Rumí · 2004
Mixtures of truncated exponentials (MTE) potentials are an alterna-tive to discretization for approx-imating probability density func-tions (PDF’s). This paper presents MTE potentials that approximate standard PDF’s and applications of these potentials for solving inference problems in hybrid Bayesian net-works.