What is: Focal Loss?

A Focal Loss function addresses class imbalance during training in tasks like object detection. Focal loss applies a modulating term to the cross entropy loss in order to focus learning on hard misclassified examples. It is a dynamically scaled cross entropy loss, where the scaling factor decays to zero as confidence in the correct class increases. Intuitively, this scaling factor can automatically down-weight the contribution of easy examples during training and rapidly focus the model on hard examples.

Formally, the Focal Loss adds a factor $(1 - p\_{t})^\gamma$ to the standard cross entropy criterion. Setting $\gamma>0$ reduces the relative loss for well-classified examples ($p\_{t}>.5$), putting more focus on hard, misclassified examples. Here there is tunable focusing parameter $\gamma \ge 0$.

${\text{FL}(p\_{t}) = - (1 - p\_{t})^\gamma \log\left(p\_{t}\right)}$