jensen_shannon_generator
def
The f-divergence generator for Jensen-Shannon divergence is f(x) = (1/2) x log x - ((x + 1)/2) log((x + 1)/2).
Statement
noncomputable def jensen_shannon_generator (x : ℝ) : ℝ
- Defined in
- arXiv:2610.21231, Section II, Example 1(3)
- Verified
- 03 Nov 2026
- Axioms
- none listed
- Used by 3
- jensen_shannon_binary_cid_not_subadditiveThere exists a binary Markov chain with conditionally identically distributed observations Y, Z given X where the unregularized…
- jensen_shannon_regularized_generatorThe regularized generator with parameter c is f_c(x) = f_{JS}(x) + c x log x.
- jensen_shannon_regularized_generator_oneFor any parameter c, the regularized generator evaluates to 0 at 1: f_c(1) = 0.
Read back from the Lean
The definition jensen_shannon_generator is a function from ℝ to ℝ mapping each real number x to (1 / 2) * x * ln(x) - ((x + 1) / 2) * ln((x + 1) / 2), where ln denotes the real natural logarithm (Real.log).
Written by a model that saw only the Lean, never the English above. If the two disagree, that is worth a challenge.
Lean source view module on GitHub
/-- The f-divergence generator for Jensen-Shannon divergence with natural logarithms. -/ noncomputable def jensen_shannon_generator (x : ℝ) : ℝ := (1 / 2) * x * Real.log x - ((x + 1) / 2) * Real.log ((x + 1) / 2)