Experience riskThe model's predictive ability for training samples is demonstrated by calculating the loss function once for all training samples and then accumulating the average, where the loss function is the basis of expected risk, empirical risk, and structural risk.

The loss function is for a single specific sample and represents the gap between the model's predicted value and the true value.

In practical applications, empirical risk minimization is usually pursued. Empirical risk is the average minimization of the loss function of all sample points in the training set. The smaller the empirical risk is, the better the model fits the training set.

References

【1】Machine Learning–> The relationship between expected risk, empirical risk and structural risk