Parameterized Temperature Scaling for Boosting the Expressive Power in Post-Hoc Uncertainty Calibration.
Christian TomaniDaniel CremersFlorian BuettnerPublished in: CoRR (2021)
Keyphrases
- expressive power
- post hoc
- first order logic
- data complexity
- query language
- computational properties
- relational algebra
- transitive closure
- learning algorithm
- propositional logic
- pattern queries
- machine learning
- databases
- incomplete information
- possibility theory
- knowledge representation
- relational calculus
- recursive queries
- aggregate functions
- data model