Single Ensemble: BLAbern

Description

BLA Bernoulli - Bayesian Learning Automaton for Bernoulli Distributed Feedback[Braadland_Norheim] An advantage of this is, that the only relevant parameters are the initial values of the beta distribution. A disadvantage is the stationarity of the algorithm. The ensemble problem in SPOT is of dynamic nature.
The default reward function is spotConfig$seq.ensemble.feed.func<-spotFeedback.reward.bern.

Usage

1
spotEnsembleSingleBLAbern(rawB, mergedB, design, spotConfig, fit = NULL)

Arguments

rawB

unmerged data

mergedB

merged data

design

new design points which should be predicted

spotConfig

global list of all options, needed to provide data for calling functions

fit

if an existing model ensemble fit is supplied, the models will not be build based on data, but only evaluated with the existing fits (on the design data). To build the model, this parameter has to be NULL. If it is not NULL the parameters mergedB and rawB will not be used at all in the function.

Details

This is a "single ensemble", meaning that in every sequential step only one model in the ensemble is trained and evaluated. The target is to actively "learn" which of the models are most suitable, based on their individual success.
The models used are specified in the spotConfig list, for instance:
spotConfig$seq.ensemble.predictors = c(spotPredictRandomForest, spotPredictEarth, spotPredictForrester, spotPredictDace)
To specify the settings of each individual model, set:
seq.ensemble.settings = list(list(setting=1),list(setting=2),list(setting=3),list(setting=4))
Any parameters set in each of the corresponding lists (here: 4 individual lists) will overwrite settings in the main spotConfig list, when the concerned model function is called.

Value

returns the list spotConfig

References

- O.-C. Granmo. A Bayesian Learning Automaton for Solving Two-Armed Bernoulli Bandit Problems. Machine Learning and Applications, ICMLA '08. p. 23-30. 2008.
- M. Friese, M. Zaefferer, T. Bartz-Beielstein, O. Flasch, P. Koch, W. Konen, and B. Naujoks. Ensemble based optimization and tuning algorithms. In F. Hoffmann and E. Huellermeier, editors, Proceedings 21. Workshop Computational Intelligence, p. 119-134. Universitaetsverlag Karlsruhe. 2011.

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.