The Setting
A binary variable can be parametrized by its bias or by its entropy deficit. The scalar rate uses the nonnegative bias, and the inverse entropy deficit expresses that rate in information coordinates.
Write for the entropy of a random sign with mean , measured in natural logarithms. The constant and the binary entropy deficit are given by (1), with : (1) The endpoint convention for the real-valued formulas here is .
The Claim
Definition 1. The inverse entropy deficit is the inverse of the restriction of to , as in (2): (2)✦ The scalar rate is the function defined by (3): (3)✦ The rate surplus is the excess of over twice the entropy deficit, given by (4): (4)✦ The entropy correction is the zero extension of the surplus in the entropy variable, defined by (5): (5)✦
Theorem 1. For an entropy deficit , the inverse identity is (6): (6)✦ For a nonnegative bias , the reverse identity is (7): (7)✦ The inverse has the continuity property (8): (8)✦ For a nonnegative bias , the rate has the parametrization (9): (9)✦ For positive entropy , the correction has the clipped form (10): (10)✦
What Is Checked
The statements concern the real binary entropy deficit restricted to nonnegative biases in , its inverse on , the associated rate and surplus, and the zero extension of the entropy correction, with the clipped formula restricted to .