Pérdida de Huber

Pérdida de Huber

Pérdida de Huber is a popular función de pérdida used in regression problems, particularly in aprendizaje automático and statistics. It combines the advantages of two other funciones de pérdida: Error cuadrático medio (MSE) and error absoluto medio (MAE). Unlike MSE, which can be heavily influenced by outliers due to the squaring of errors, Huber Loss is designed to be robust against such anomalies.

La pérdida de Huber se define mediante un parámetro llamado umbral (a menudo denotado como δ), which determines the point at which the loss function transitions from quadratic to linear. For residuals (the differences between actual and predicted values) that are less than δ en valor absoluto, la pérdida de Huber se comporta como MSE, usando la fórmula:

Pérdida de Huber = 0.5 * (residual)^2

Para residuos que superan δ, the loss is calculated using the absolute error formula, which is less sensitive to large errors:

Huber Loss = δ * (|residual| – 0.5 * δ)

Esta combinación permite que la pérdida de Huber proporcione un gradiente suave para optimization while limiting the influence of outliers. When selecting δ, it is important to consider the scale of the data and the specific characteristics of the dataset.

Huber Loss is particularly useful in scenarios where a dataset contains outliers that could skew the results if MSE were used exclusively. It strikes a balance between maintaining sensitivity to small errors and robustness against large deviations, making it a versatile choice for many regression applications.