p

com.cra.figaro.experimental

collapsedgibbs

package collapsedgibbs

Ordering
  1. Alphabetic
Visibility
  1. Public
  2. All

Type Members

  1. abstract class CollapsedProbQueryGibbs extends ProbQueryGibbs with CollapsedProbabilisticGibbs

    CollapsedProbQueryGibbs only uses graph information and the list of targets to collapse some variables.

    CollapsedProbQueryGibbs only uses graph information and the list of targets to collapse some variables. extend with HeuristicCollapser or RecurringCollapser to implement other features described in Gogate et. al.

  2. trait CollapsedProbabilisticGibbs extends BaseUnweightedSampler with ProbabilisticGibbs
  3. trait DeterministicCollapseStrategy extends BaseUnweightedSampler with CollapsedProbabilisticGibbs

    Experimental collapser that doesn't need to calculate marginal probabilities.

    Experimental collapser that doesn't need to calculate marginal probabilities. The original paper doesn't distinguish between model variables, or use any meta-information about the variables. Since Figaro knows which variables are deterministic, we can use this as a proxy for the correlation heuristic.

  4. trait FactorSizeCollapseStrategy extends BaseUnweightedSampler with CollapsedProbabilisticGibbs

    This trait causes variables to collapsed until the total summed size of all of the factors collapsed thus far exceeds a threshold.

  5. trait HeuristicCollapseStrategy extends BaseUnweightedSampler with CollapsedProbabilisticGibbs

    HeuristicCollapsedGibbs adds the Hellinger-distance-based term to the elimination heuristic.

    HeuristicCollapsedGibbs adds the Hellinger-distance-based term to the elimination heuristic. So the heuristic is now based on the marginal probabilities and pairwise marginals. These have to be estimated by some number of saved samples.

  6. trait RecurringCollapseStrategy extends BaseUnweightedSampler with HeuristicCollapseStrategy

    In the paper, the authors recommend updating the marginals every N samples and re-collapsing every few iterations.

    In the paper, the authors recommend updating the marginals every N samples and re-collapsing every few iterations. In practice, this is pretty slow. This trait will keep a running tally of samples of each of the used variables and re-collapse the factor graph (starting from the initial graph) periodically.

Value Members

  1. object CollapsedGibbs

Ungrouped