scientific article; zbMATH DE number 3991410
[For the entire collection see Zbl 0595.00024.] The relative entropy is an information-theoretic measure of the dissimilarity between two probability distributions. Various properties of relative entropy have led to its widespread use in information theory. These properties suggest that relative entropy has a role to play in systems that attempt to perform inference in terms of probability distributions. The paper reviews some basic properties of relative entropy as well as its role in probabilistic inference. The paper points out three areas in artificial intelligence (AI), where relative entropy should have a role to play. 1. Quantifying information gains and losses. 2. Probabilistic inferencing. 3. Searching large state spaces.
- Characterizing the principle of minimum cross-entropy within a conditional-logical framework
- Information and probabilistic reasoning
- SPIRIT and Léa Sombé: A study in probabilistic reasoning
- Explaining default intuitions using maximum entropy.
- A note on conditional logics and entropy
- Conditional logic and the Principle of Entropy
- Entropy and monotonicity in artificial intelligence
- A conditional-logical approach to minimum cross-entropy
- Attention to entropic communication
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3753919)