Results 1  10
of
4,009
Bounds on marginal probability distributions
 Advances in Neural Information Processing Systems 21 (NIPS*2008
, 2008
"... We propose a novel bound on singlevariable marginal probability distributions in factor graphs with discrete variables. The bound is obtained by propagating local bounds (convex sets of probability distributions) over a subtree of the factor graph, rooted in the variable of interest. By constructio ..."
Abstract

Cited by 16 (0 self)
 Add to MetaCart
We propose a novel bound on singlevariable marginal probability distributions in factor graphs with discrete variables. The bound is obtained by propagating local bounds (convex sets of probability distributions) over a subtree of the factor graph, rooted in the variable of interest
Learning unbelievable marginal probabilities
"... Loopy belief propagation performs approximate inference on graphical models with loops. One might hope to compensate for the approximation by adjusting model parameters. Learning algorithms for this purpose have been explored previously, and the claim has been made that every set of locally consist ..."
Abstract
 Add to MetaCart
consistent marginals can arise from belief propagation run on a graphical model. On the contrary, here we show that many probability distributions have marginals that cannot be reached by belief propagation using any set of model parameters or any learning algorithm. We call such marginals â€˜unbelievable
Transitive Comparison of Marginal Probability Distributions
"... A generalized dice model for the pairwise comparison of nonnecessarily independent random variables is established. It is shown how the transitivity of the probabilistic relation generated by the model depends on the copula defining the coupling of the marginal distribution functions in the joint d ..."
Abstract
 Add to MetaCart
A generalized dice model for the pairwise comparison of nonnecessarily independent random variables is established. It is shown how the transitivity of the probabilistic relation generated by the model depends on the copula defining the coupling of the marginal distribution functions in the joint
Deterministic Approximation of Marginal Probabilities in Bayes Nets
, 1998
"... Computation of marginal probabilities in Bayes nets is central to numerous reasoning and automatic decisionmaking systems. This paper presents a deterministic approximation scheme for this hard problem, that supplies provably correct bounds, by aggregating probability mass in IndependenceBased (IB ..."
Abstract

Cited by 6 (2 self)
 Add to MetaCart
Computation of marginal probabilities in Bayes nets is central to numerous reasoning and automatic decisionmaking systems. This paper presents a deterministic approximation scheme for this hard problem, that supplies provably correct bounds, by aggregating probability mass in Independence
Exploiting CaseBased Independence for Approximating Marginal Probabilities
 International Journal of Approximate Reasoning
, 1994
"... Computing marginal probabilities (whether prior or posterior) in Bayesian belief networks is a hard problem. This paper discusses deterministic approximation schemes that work by adding up the probability mass in a small number of value assignments to the network variables. Under certain assumptions ..."
Abstract

Cited by 14 (7 self)
 Add to MetaCart
Computing marginal probabilities (whether prior or posterior) in Bayesian belief networks is a hard problem. This paper discusses deterministic approximation schemes that work by adding up the probability mass in a small number of value assignments to the network variables. Under certain
COMPARING VARIOUS MARGINAL PROBABILITY MODELS IN EVOLUTIONARY ALGORITHMS
"... Evolutionary algorithms based on probabilistic modeling do not use genetic operators anymore. Instead, they learn a probabilistic model and create new population of potential solutions by sampling from this model. This paper presents some results of an empirical comparison of this type of evolutiona ..."
Abstract
 Add to MetaCart
of evolutionary algorithm using various marginal probability models (three types of histograms and gaussian mixture) in continuous domain. We found that the equiheight histogram and maxdiff histogram are the winners for the type of evaluation functions considered here, although the mixture of gaussians offers
Solving security games on graphs via marginal probabilities
, 2013
"... Security games involving the allocation of multiple security resources to defend multiple targets generally have an exponential number of pure strategies for the defender. One method that has been successful in addressing this computational issue is to instead directly compute the marginal probabili ..."
Abstract

Cited by 10 (3 self)
 Add to MetaCart
probabilities with which the individual resources are assigned (first pursued by Kiekintveld et al. (2009)). However, in sufficiently general settings, there exist games where these marginal solutions are not implementable, that is, they do not correspond to any mixed strategy of the defender. In this paper, we
ARMA Signals with Specified Symmetric Marginal Probability Distribution
"... Except in the case of normal (i.e Gaussian) distribution, it is very difficult to calculate the marginal probability distribution of ARMA signals. By using a particular form of modeling such signals with random coefficients we show that the problem can be solved and we present an algorithm allowing ..."
Abstract
 Add to MetaCart
Except in the case of normal (i.e Gaussian) distribution, it is very difficult to calculate the marginal probability distribution of ARMA signals. By using a particular form of modeling such signals with random coefficients we show that the problem can be solved and we present an algorithm allowing
Exploiting IB Assignments for Approximating Marginal Probabilities
, 1994
"... Computing marginal probabilities (whether prior or posterior) in Bayesian belief networks is a hard problem. This paper discusses deterministic approximation schemes that work by adding up the probability mass in a small number of value assignments to the network variables. Under certain assumptions ..."
Abstract
 Add to MetaCart
Computing marginal probabilities (whether prior or posterior) in Bayesian belief networks is a hard problem. This paper discusses deterministic approximation schemes that work by adding up the probability mass in a small number of value assignments to the network variables. Under certain
Results 1  10
of
4,009