gradientrelated
gradientrelated is a label used in mathematics, computer science, and related disciplines to refer broadly to objects, methods, or analyses that depend on or exploit gradient information. In calculus, the gradient of a function defines its rate of change and the direction of steepest ascent; gradientrelated concepts study or utilize this information to derive updates, assess sensitivity, or model dynamical behavior.
In optimization, gradientrelated methods include Gradient Descent, Gradient Ascent, and their variants such as stochastic gradient
In machine learning and deep learning, backpropagation computes gradients of a loss with respect to parameters,
The term gradientrelated is informal and tends to appear in overview texts rather than formal definitions.