Loss Function

The loss function quantifies the difference between the predicted output and the actual output, guiding the training process by informing the model how well it is performing. Common loss functions include mean squared error for regression and cross-entropy for classification tasks. Minimizing the loss function is the goal during training.