ConceptAI & ML
Huber loss
A loss function that interpolates between a squared error loss for small errors and an absolute value loss for large errors to prevent excessively large gradients.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Lecturers
1Connections
3The lectures that define it or assume it, and the concepts next to it.