Section d, p.8 — Toussaint's inequality
ProvedMayersBC.toussaintcryptographyp2o-gran-per-chapterp2o-plan-paperp2o-strengthenedp2o-v1quantum-information
Let be probability distributions on a finite set , and let
Then
Mayers applies it to the outcome distributions of a measurement on : together with it gives .
Differences from the source:
- generalization: BW and K of the outcome distributions Pr(J_0 = j), Pr(J_1 = j) of a measurement → any two probability distributions on a finite set; reason: the inequality cited from [15] holds for all distributions; outcome distributions are a special case
- encoding: BW = Σ_j Pr(J_0 = j)^{1/2} Pr(J_1 = j)^{1/2} → FuchsVdGraaf.bhattacharyya p0 p1 = Σ_j √(p0(j) p1(j)); reason: the two expressions agree for nonnegative values
Preamble
import Mathlib import Definitions.Def_FuchsVdGraaf_IsProbDist import Definitions.Def_FuchsVdGraaf_bhattacharyya import Definitions.Def_MayersBC_kolmogorov
Formal statement
namespace MayersBC
/-- Mayers, arXiv:quant-ph/9603015v3, section d, p.8: "It is explained in [15] that
(1 − BW) ≤ K/2", where `BW = Σ_{j∈A} Pr(J_0 = j)^{1/2} Pr(J_1 = j)^{1/2}` is the
Bhattacharyya–Wootters overlap and `K = Σ_j |p_0(j) − p_1(j)|` the Kolmogorov distance (no ½).
For any two probability distributions `p_0, p_1` on a finite set, `1 − BW(p_0, p_1) ≤ K(p_0, p_1)/2`.
**Formalization Note.** `BW` is the published `FuchsVdGraaf.bhattacharyya`,
`Σ_j √(p_0(j) p_1(j))`, which equals `Σ_j √p_0(j) √p_1(j)` for nonnegative values. Distributions
are the published `FuchsVdGraaf.IsProbDist`. -/
theorem toussaint {J : Type} [Fintype J] (p0 p1 : J → ℝ)
(h0 : FuchsVdGraaf.IsProbDist p0) (h1 : FuchsVdGraaf.IsProbDist p1) :
1 - FuchsVdGraaf.bhattacharyya p0 p1 ≤ kolmogorov p0 p1 / 2 := by sorry
end MayersBC
Source
Mayers, The Trouble with Quantum Bit Commitment, arXiv:quant-ph/9603015v3, section d, p.8
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.