Faddeev’s axioms

E45254

Faddeev’s axioms are a set of conditions characterizing Shannon entropy in information theory, providing an alternative but equivalent axiomatization to the Shannon–Khinchin framework.


Statements (30)

Predicate Object
instanceOf axiomatization of entropy
characterization of Shannon entropy
set of axioms in information theory
appliesTo finite discrete probability distributions
assumes entropy is a real-valued function on discrete probability distributions
characterizes Shannon entropy
clarifies why Shannon entropy has its specific functional form
field information theory
implies Shannon entropy is the unique solution up to scale
includesCondition continuity of entropy
normalization condition for entropy
recursivity (grouping) condition
symmetry of entropy under permutation of outcomes
influenced later axiomatizations of entropy and information measures
isAlternativeTo Shannon–Khinchin axioms
isEquivalentTo Shannon–Khinchin axioms
mainGoal uniquely characterize Shannon entropy up to a positive multiplicative constant
mathematicalDomain functional equations
probability theory
namedAfter Dmitry Faddeev
relatedTo Shannon entropy
Shannon–Khinchin axioms
characterization theorems for entropy
requires entropy of a certain two-point distribution to be positive
supports uniqueness of Shannon entropy under natural requirements
typeOfCondition axiomatic characterization
usedFor foundations of information theory
justification of Shannon entropy
usesConcept information measure
probability distribution

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Shannon–Khinchin axioms relatedTo Faddeev’s axioms