.. _method-bayes_calibration-gpmsa:
"""""
gpmsa
"""""
(Experimental) Gaussian Process Models for Simulation Analysis
(GPMSA) Bayesian calibration
**Topics**
package_queso, bayesian_calibration
.. toctree::
:hidden:
:maxdepth: 1
method-bayes_calibration-gpmsa-chain_samples
method-bayes_calibration-gpmsa-seed
method-bayes_calibration-gpmsa-rng
method-bayes_calibration-gpmsa-build_samples
method-bayes_calibration-gpmsa-import_build_points_file
method-bayes_calibration-gpmsa-standardized_space
method-bayes_calibration-gpmsa-logit_transform
method-bayes_calibration-gpmsa-gpmsa_normalize
method-bayes_calibration-gpmsa-export_chain_points_file
method-bayes_calibration-gpmsa-dram
method-bayes_calibration-gpmsa-delayed_rejection
method-bayes_calibration-gpmsa-adaptive_metropolis
method-bayes_calibration-gpmsa-metropolis_hastings
method-bayes_calibration-gpmsa-proposal_covariance
method-bayes_calibration-gpmsa-options_file
**Specification**
- *Alias:* None
- *Arguments:* None
**Child Keywords:**
+-------------------------+--------------------+------------------------------+---------------------------------------------+
| Required/Optional | Description of | Dakota Keyword | Dakota Keyword Description |
| | Group | | |
+=========================+====================+==============================+=============================================+
| Required | `chain_samples`__ | Number of Markov Chain Monte Carlo |
| | | posterior samples |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `seed`__ | Seed of the random number generator |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `rng`__ | Selection of a random number generator |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Required | `build_samples`__ | Number of initial model evaluations used in |
| | | build phase |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `import_build_points_file`__ | File containing points you wish to use to |
| | | build a surrogate |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `standardized_space`__ | Perform Bayesian inference in standardized |
| | | probability space |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `logit_transform`__ | Utilize the logit transformation to reduce |
| | | sample rejection for bounded domains |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `gpmsa_normalize`__ | Enable GPMSA-internal normalization |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `export_chain_points_file`__ | Export the MCMC chain to the specified |
| | | filename |
+-------------------------+--------------------+------------------------------+---------------------------------------------+
| Optional (Choose One) | MCMC Algorithm | `dram`__ | Use the DRAM MCMC algorithm |
| | +------------------------------+---------------------------------------------+
| | | `delayed_rejection`__ | Use the Delayed Rejection MCMC algorithm |
| | +------------------------------+---------------------------------------------+
| | | `adaptive_metropolis`__ | Use the Adaptive Metropolis MCMC algorithm |
| | +------------------------------+---------------------------------------------+
| | | `metropolis_hastings`__ | Use the Metropolis-Hastings MCMC algorithm |
+-------------------------+--------------------+------------------------------+---------------------------------------------+
| Optional | `proposal_covariance`__ | Defines the technique used to generate the |
| | | MCMC proposal covariance. |
+----------------------------------------------+------------------------------+---------------------------------------------+
| Optional | `options_file`__ | File containing advanced QUESO/GPMSA |
| | | options |
+----------------------------------------------+------------------------------+---------------------------------------------+
.. __: method-bayes_calibration-gpmsa-chain_samples.html
__ method-bayes_calibration-gpmsa-seed.html
__ method-bayes_calibration-gpmsa-rng.html
__ method-bayes_calibration-gpmsa-build_samples.html
__ method-bayes_calibration-gpmsa-import_build_points_file.html
__ method-bayes_calibration-gpmsa-standardized_space.html
__ method-bayes_calibration-gpmsa-logit_transform.html
__ method-bayes_calibration-gpmsa-gpmsa_normalize.html
__ method-bayes_calibration-gpmsa-export_chain_points_file.html
__ method-bayes_calibration-gpmsa-dram.html
__ method-bayes_calibration-gpmsa-delayed_rejection.html
__ method-bayes_calibration-gpmsa-adaptive_metropolis.html
__ method-bayes_calibration-gpmsa-metropolis_hastings.html
__ method-bayes_calibration-gpmsa-proposal_covariance.html
__ method-bayes_calibration-gpmsa-options_file.html
**Description**
GPMSA (Gaussian Process Models for Simulation Analysis) is a
surrogate-based Markov Chain Monte Carlo Bayesian calibration
method. Dakota's GPMSA is an experimental capability and not ready for
production use at this time.
Central to GPMSA is the construction of a Gaussian Process emulator
from simulation runs collected at various settings of input
parameters. The emulator is a statistical model of the system
response, and it is used to incorporate the observational data to
improve system predictions and constrain or calibrate the unknown
parameters. The GPMSA code draws heavily on the theory developed in
the seminal Bayesian calibration paper by Kennedy and O'Hagan
:cite:p:`Kenn01`. The particular approach in GPMSA was developed by the Los
Alamos National Labortory statistics group and documented in :cite:p:`Hig08`.
Dakota's GPMSA capability comes from the QUESO package developed at UT Austin.
*Usage Tips:*
Configuring GPMSA essentially involves identifying the simulation
build data, the experiment data, the calibration and configuration
(state) variables, and any necessary algorithm controls. The GP
surrogate model is automatically constructed internal to the algorithm
and need not be specified through Dakota input.
Dakota's GPMSA implementation is not intended for production use.
There are a number of known limitations, including:
- Only works for scalar and multivariate responses, not field responses. Field responses will be treated as a single multi-variate response set. Consequently, simulation and experiment data must have the same dimensions.
- When build data is not imported a design of experiments will be conducted over all calibration and scenario variables present.
- Experiment data is required (one cannot pose the simulation data as a set of residuals with the assumption of 0-valued experiments).
- Output and diagnostics are limited. Advanced users will need to examine QUESO output files (potentially written in a transformed scaled space) in the QuesoDiagnostic directory
**Examples**
The following input file fragment illustrates GPMSA-based Bayesian
calibration of 3 :math:`\beta` variables with a uniform prior, with 3
configuration (scenario) variables :math:`x` . A total of 60 simulation
build points are provided in sim_data.dat, which contains
columns for each :math:`\beta` , followed by each :math:`x` , and then the
simulation response 'lin'. Each row of the experiment data file
y_exp_with_var.dat contains the values of the 3 :math:`x`
variables, followed by the value of 'lin' and its observation error
(variance).
.. code-block::
method
bayes_calibration gpmsa
chain_samples 1000
seed 2460
build_samples 60
import_build_points_file 'sim_data.dat' freeform
export_chain_points_file 'posterior.dat'
burn_in_samples = 100 sub_sampling_period = 2
posterior_stats kl
variables
uniform_uncertain 3
upper_bounds 0.4500 -0.1000 0.4000
initial_point 0.2750 -0.3000 0.1000
lower_bounds -0.1000 -0.5000 -0.2000
descriptors 'beta0' 'beta1' 'beta2'
continuous_state 3
upper_bounds 3 * 1.0
initial_state 3 * 0.5
lower_bounds 3 * 0.0
descriptors 'x0' 'x1' 'x2'
responses
descriptors 'lin'
calibration_terms 1
calibration_data_file 'y_exp_with_var.dat'
freeform
num_experiments 5
num_config_variables 3
experiment_variance_type 'scalar'
no_gradients
no_hessians