Techniques for automated bayesian posterior sampling using markov chain monte carlo and related schemes
AssigneeSAS Institute Inc.
InventorsChristian Macaro, Jan Chvosta, Mark Roland Little
Life of this patent
Estimating the term…
Abstract
Techniques for automated Bayesian posterior sampling using Markov Chain Monte Carlo and related schemes are described. In an embodiment, one or more values in a stationarity phase for a system configured for Bayesian sampling may be initialized. Sampling may be performed in the stationarity phase based upon the one or more values to generate a plurality of samples. The plurality of samples may be evaluated based upon one or more stationarity criteria. The stationarity phase may be exited when the plurality of samples meets the one or more stationarity criteria. Other embodiments are described and claimed.
Claims28
An article of manufacture comprising a non-transitory computer-readable storage medium comprising instructions that, when executed, cause a system to:
- initialize one or more values in a stationarity phase for a system configured for Bayesian sampling;
- perform sampling in the stationarity phase based upon the one or more values to generate a plurality of samples;
- evaluate the plurality of samples based upon one or more stationarity criteria;
- and
- exit the stationarity phase when the plurality of samples meets the one or more stationarity criteria.
A computer-implemented method, comprising:
- initializing, with a computing system configured for Bayesian sampling, one or more values in a stationarity phase;
- performing sampling in the stationarity phase based upon the one or more values to generate a plurality of samples;
- evaluating the plurality of samples based upon one or more stationarity criteria;
- and
- exiting the stationarity phase when the plurality of samples meets the one or more stationarity criteria.
An apparatus, comprising:
- processor circuitry;
- and
- at least one memory unit communicatively coupled to the processor circuitry, the at least one memory unit configured to store:
- an initialization module configured to initialize one or more values in a stationarity phase for a system configured for Bayesian sampling;
- a sampling module configured to perform sampling in the stationarity phase based upon the one or more values to generate a plurality of samples;
- and
- an evaluation module configured to evaluate the plurality of samples based upon one or more stationarity criteria, and exit the stationarity phase when the plurality of samples meets the one or more stationarity criteria.
A computer-program product tangibly embodied in a non-transitory machine-readable storage medium, the computer-program product including instructions operable to cause a computing device to perform operations including:
- executing a stationarity phase of a system configured for Bayesian sampling, the execution of the stationarity phase comprising operations including:
- initialize one or more values;
- perform sampling in the stationarity phase based upon the one or more values to generate a plurality of samples;
- accessing one or more stationarity criteria;
- evaluate the plurality of samples based upon the one or more stationarity criteria;
- and
- ending the execution of the stationarity phase when the plurality of samples meets the one or more stationarity criteria.
Dependencies are inferred from the claim text.