In the context of loss functions, the _____ is a method used to minimize the difference between predicted values and actual values by adjusting model parameters.

In the context of loss functions, the gradient descent is a method used to minimize the difference between predicted values and actual values by adjusting model parameters. It works by computing the gradient of the loss with respect to each parameter, which tells how the loss changes when that parameter changes. The algorithm then moves each parameter slightly in the direction that reduces the loss, repeating this process until the loss stops decreasing. For example, if a model predicts 8 but the true value is 10, gradient descent will adjust the weights so that future predictions move closer to 10. This iterative adjustment is what makes the model learn from its mistakes.