Towards a More Efficient Computation of Weighted Conditional Impacts for Relational Probabilistic Knowledge Bases Under Maximum Entropy Semantics.
Marc FinthammerChristoph BeierlePublished in: KI (2015)
Keyphrases
- efficient computation
- maximum entropy
- probabilistic logic
- maximum entropy principle
- default reasoning from conditional knowledge bases
- principle of maximum entropy
- knowledge base
- maximum entropy model
- probabilistic knowledge
- logic programming
- computational efficiency
- description logics
- transformation based learning
- conditional random fields
- knowledge representation
- data model
- generative model
- probabilistic model
- probabilistic reasoning
- iterative scaling
- databases
- image segmentation
- probability theory
- default reasoning
- belief revision
- multi class