ML architecture that utilizes multiple specialist models (experts) to handle different parts of the input space, coordinated by a gating mechanism that decides which expert to use for each input.
Generality: 705
