Rejection sampling of a complex distribution including bound and proposal distribution refinement
AssigneeXerox Corporation
InventorsMarc Dymetman, Guillaume M. Bouchard
Life of this patent
Estimating the term…
Loading patent…
AssigneeXerox Corporation
InventorsMarc Dymetman, Guillaume M. Bouchard
Estimating the term…
Iterative rejection sampling is performed on a domain in accordance with a target distribution. The domain is partitioned to define a partition comprising partition elements, and each iteration of the rejection sampling includes selecting a partition element from the partition in accordance with partition element selection probabilities. A sample of the domain is acquired in the selected partition element according to a normalized proposal distribution that is associated with and normalized over the selected partition element. The acquired sample is accepted or rejected based on the target distribution and a bound associated with the selected partition element. During the iterative rejection sampling, the partition is adapted by replacing a partition element of the partition with two or more split partition elements, associating bounds with the split partition elements, and computing partition element selection probabilities for the split partition elements.
A method comprising:
A non-transitory storage medium storing instructions executable by an electronic processing device to perform a method including performing iterative rejection sampling of a domain in accordance with a target distribution p wherein the domain is partitioned into a partition comprising partition elements Yi each having an associated selection probability πi and an associated normalized proposal distribution qYi that is normalized over the partition element Yi and an associated bound QYi=BYi qYi where BYi is a nonnegative scaling factor, wherein at least one iteration of the iterative rejection sampling is performed by:
An apparatus comprising:
Dependencies are inferred from the claim text.