Identification of out-of-distribution cases of CNN using class-based surprise adequacy.
Mira MarhabaEttore MerloFoutse KhomhGiuliano AntoniolPublished in: CAIN (2022)
Keyphrases
- mild assumptions
- cellular neural networks
- automatic identification
- case base
- real time
- case based reasoning
- similarity measure
- case study
- information systems
- databases
- probability distribution
- d objects
- random variables
- data distribution
- image processing
- class specific
- learning algorithm
- class conditional
- class membership