What does “Loss Function” mean?

The model error value computed during training and optimized downward over time.

A common objective function used to train classification models by penalizing incorrect probabilities.

The core training algorithm that updates model weights by propagating prediction errors backward through the network.

A training hyperparameter controlling how much parameters change each update step.

Using a language model to score or compare outputs from other models during evaluation.

A consistent pattern of error or unfairness in data or model behavior.