Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 2.4, p. 211 — if g meets every eigenspace of H and exact CG ends with r = 0, then H is positive definite

Proved
TruncNewton.NegCurv.pd_of_cg_residual_zero

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

conjugate-gradienteigenspacenegative-curvaturep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1positive-definite

Let HHH be a symmetric linear operator on Rn\mathbb{R}^nRn and g∈Rng\in\mathbb{R}^ng∈Rn, and run the TNCG minor iteration (Steps 1–4, p. 194) with ε=η=0\varepsilon=\eta=0ε=η=0, producing directions djd_jdj​ and residuals rjr_jrj​. Suppose that ggg is not orthogonal to any eigenspace of HHH, i.e. for every eigenvalue μ\muμ of HHH there is vvv with Hv=μvHv=\mu vHv=μv and gTv≠0g^{\mathsf T}v\ne0gTv=0. Suppose further that for some i≥0i\ge0i≥0 the curvature test (2.1) with ε=0\varepsilon=0ε=0 passes at every iteration up to iii,

djTHdj>0,j=0,1,…,i,d_j^{\mathsf T}Hd_j>0,\qquad j=0,1,\dots,i,djT​Hdj​>0,j=0,1,…,i,

and that the iteration then ends with zero residual, ri+1=0r_{i+1}=0ri+1​=0 (the test (2.2) with η=0\eta=0η=0). Then HHH is positive definite:

vTHv>0for all v≠0.v^{\mathsf T}Hv>0\qquad\text{for all } v\ne0 .vTHv>0for all v=0.

This is the first sentence of the proof of Theorem 2.4, obtained there from Theorem A.5 together with (A.2) and (A.5); it says that exact CG can only end with r=0r=0r=0 at a non-convex point when ggg misses some eigenspace.

Formalization Note "r=0r=0r=0" is encoded as ri+1=0r_{i+1}=0ri+1​=0 after i+1i+1i+1 CG steps whose curvature was positive at every step, which is exactly the exit through (2.2) with η=0\eta=0η=0 at iteration iii. HHH symmetric is the paper's standing setting (HHH is a Hessian).

Preamble
import Mathlib
import Definitions.Def_TruncNewton_NegCurv_Setting
Formal statement
namespace TruncNewton.NegCurv

theorem pd_of_cg_residual_zero {n : ℕ}
    (H : EuclideanSpace ℝ (Fin n) →L[ℝ] EuclideanSpace ℝ (Fin n)) (hH : IsSelfAdjoint H)
    (g : EuclideanSpace ℝ (Fin n))
    (hg : ∀ μ : ℝ, (∃ v, v ≠ 0 ∧ H v = μ • v) → ∃ v, H v = μ • v ∧ inner ℝ g v ≠ 0)
    (i : ℕ) (hcurv : ∀ j ≤ i, 0 < inner ℝ ((cg H g j).d) (H (cg H g j).d))
    (hr : (cg H g (i + 1)).r = 0) :
    ∀ v, v ≠ 0 → 0 < inner ℝ v (H v) := by sorry

end TruncNewton.NegCurv
Source
Dembo and Steihaug, Truncated-Newton algorithms for large-scale unconstrained optimization, Math. Programming 26 (1983), p. 211, proof of Theorem 2.4, first sentence
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me