Attribute sampling is a training technique in which a decision forest or similar ensemble considers only a random subset of features when selecting split conditions. By varying the available features across nodes or trees, it can reduce correlation among trees and improve generalization.
A tactic for training a decision forest in which each decision tree considers only a random subset of possible features when learning the condition. Generally, a different subset of features is sampled for each node. In contrast, when training a decision tree without attribute sampling, all possible features are considered for each node.