empirical risk minimization (ERM)

Empirical risk minimization (ERM) is a principle where the model is trained to minimize the average loss (empirical risk) over the training dataset. The empirical risk is calculated as the average of the loss function over all examples in the training set. By minimizing this risk, the model learns to make predictions that closely match the observed data. However, the choice of the loss function and the method of minimization can significantly impact the model's performance and its ability to generalize to new, unseen data.