EDISON (version 1.1.1)

ExpHyperRatioTarget: Calculates the ratio of an exponential hyperparameter move.

Description

This function calculates the acceptance ratio of a level-1 hyperparameter move for a given target node.

Usage

ExpHyperRatioTarget(beta.proposed, beta.old, target.net, self.loops)

Arguments

beta.proposed

Proposed new hyperparameter value.

beta.old

Previous value of hyperparameter beta.

target.net

Network segments for the target node associated with this hyperparameter value.

self.loops

'TRUE' if self-loops are acceptable, 'FALSE' otherwise.

Value

Returns the ratio of the exponential prior with the previous hyperparameter value and the proposed new hyperparameter value.

References

For information about the exponential information sharing prior, see:

Husmeier et al. (2010), "Inter-time segment information sharing for non-homogeneous dynamic Bayesian networks", NIPS.

Dondelinger et al. (2012), "Non-homogeneous dynamic Bayesian networks with Bayesian regularization for inferring gene regulatory networks with gradually time-varying structure", Machine Learning.

See Also

ExpHyperMove