A foundational principle in statistics and ML (Machine Learning), focused on minimizing the average of the loss function over a sample dataset.
Generality: 814
