Techniques designed to reduce the size of a machine learning model without significantly sacrificing its accuracy.
Generality: 715
