A loss function for making the model focus on hard training examples.
Focal Loss is like a batting coach at practice. It stops cheering the kid hitting beach balls and helps the kid missing every pitch.
It makes easy training examples count less. You meet it in object detection. Background boxes can drown out real objects.
Cross-Entropy Loss
Focal Loss reweights Cross-Entropy Loss to care less about easy examples.
Object Detection
Focal Loss helps Object Detection handle too many background examples.
Classification
Focal Loss still improves training for Classification.