Remark 3, p. 19 — at γ = 0.911 the large edges L of Algorithm 4 form a matching
ProvedQMCProduct.HalfApprox.remark_3matchingmonogamy-of-entanglementp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1quantum-max-cut
Let be a graph on qubits, a 2-positive pseudo-density with edge values , and . The set of large edges
is a matching: two distinct edges of share no endpoint.
This is what makes the output (5) of Algorithm 4 a product of states on disjoint qubits, and it is used in the analysis of edges adjacent to .
Formalization Note Each undirected edge of is represented once, as an ordered pair with ; the conclusion states that neither endpoint of one pair equals either endpoint of the other. Only 2-positivity of is assumed, not optimality, since the remark follows from Fact 2.
Preamble
import Definitions.Def_QMCProduct_HalfApprox_Algorithm set_option autoImplicit false
Formal statement
namespace QMCProduct.HalfApprox
/-- Remark 3, p. 19: at `γ = 0.911` the large edges of Algorithm 4 form a matching. -/
theorem remark_3 {n : ℕ} (G : SimpleGraph (Fin n)) [DecidableRel G.Adj]
(ρt : Op n) (hρ : IsPseudoDensity 2 ρt)
(e f : Fin n × Fin n) (he : e ∈ large G ρt (911 / 1000 : ℝ))
(hf : f ∈ large G ρt (911 / 1000 : ℝ)) (hef : e ≠ f) :
e.1 ≠ f.1 ∧ e.1 ≠ f.2 ∧ e.2 ≠ f.1 ∧ e.2 ≠ f.2 := by sorry
end QMCProduct.HalfApprox
Source
Parekh & Thompson, An Optimal Product-State Approximation for 2-Local Quantum Hamiltonians with Positive Terms, arXiv:2206.08342v2, p. 19, Remark 3
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.