Uncertainty-based out-of-distribution detection requires suitable function space priors.
Francesco D'AngeloChristian HenningPublished in: CoRR (2021)
Keyphrases
- distribution function
- object detection
- detection rate
- dual space
- low dimensional
- prior knowledge
- search space
- automatic detection
- higher dimensional
- probability distribution
- detection method
- random variables
- density function
- computer vision
- robust optimization
- reproducing kernel hilbert space
- reinforcement learning
- decision theory
- probability density
- detection accuracy
- bayesian framework
- neural network
- space time
- particle filter