The process of adjusting the parameters of a model to minimize a loss function that measures the difference between the model’s predictions and the true values. Example: Optimizing a neural network’s parameters using a gradient descent algorithm.
Optimization
Published
· Updated





