ConceptAI & ML
Output entropy uncertainty
Using the entropy of the output distribution of a neural network to measure model uncertainty, which fails because it captures aleatoric noise rather than epistemic model error.
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.