§5, p. 22 — R̃(k) = V(k+1)⁻¹C(k)V(k) is a row stochastic matrix
ProvedDIGing.Push.rt_row_stochasticdirected-graphsp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1push-sumstochastic-matrices
Let the mixing matrices satisfy Assumption 7, let , , , and . Then for every , is row stochastic: for all
Row stochasticity of is what lets the recursion (48) be analysed like DIGing: multiplication by maps consensual matrices to themselves.
Formalization Note The paper cites Lemma 4 of [41] and says "every row sums to 1"; the nonnegativity of the entries, which is part of "row stochastic", is included. Assumption 7 includes the disclosed self-weight; Assumption 6 is not needed.
Preamble
import Mathlib import Definitions.Def_DIGing_Undir_Common import Definitions.Def_DIGing_Push_Setting
Formal statement
namespace DIGing.Push
theorem rt_row_stochastic {n : ℕ} (A : ℕ → Finset (Fin n × Fin n))
(C : ℕ → Matrix (Fin n) (Fin n) ℝ) (h7 : Assumption7 A C) :
∀ k i, (∀ j, 0 ≤ Rt C k i j) ∧ ∑ j, Rt C k i j = 1 := by sorry
end DIGing.Push
Source
arXiv:1607.03218v3, §5, p. 22 (sentence after (49)); proof of Lemma 13, p. 23
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.