this is an idea to make oblique RFs more explainable for a single prediction:
pull out the regression coefficients from the node directly above the given obervation's predicted leaf node of each tree
aggregate them
use the aggregate regression coefficients as a starting value in a regression model fitted to the training set
(optional) weight the training set by nearest neighbors of the observation
We would assess the validity of this by measuring correlation between the forest's predictions and the model's predictions. It would necessitate fitting 1 model per prediction, so not computationally great.
this is an idea to make oblique RFs more explainable for a single prediction:
We would assess the validity of this by measuring correlation between the forest's predictions and the model's predictions. It would necessitate fitting 1 model per prediction, so not computationally great.