Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 4.11 (companion theorem; not used in the proof of the goal) — variable-arity register algorithm

Open
CatalyticTreeEval.Main.Supplement.theorem_4_11

by mikedeng1 · Nov 15, 2026 · Mathlib 0df444a (Lean v4.33.1)

complexity-theoryp2o-batch-stoc27sp2o-gran-per-chapterp2o-plan-paperp2o-v1tree-evaluation

In the arity-rrr framework, for every integer r≥2r\ge2r≥2, take m>1m>1m>1, a positive label length ℓ\ellℓ, an rrr-noncancelling score set SSS containing zero, an SSS-matching family of exact size 2ℓ2^\ell2ℓ and dimension ddd, and an rSrSrS-decoder of sparsity kkk. Let fixed uniform access procedures have common space s≥log⁡2(dm)s\ge\log_2(dm)s≥log2​(dm). A single deterministic binary machine evaluates every complete ordered rrr-ary tree of every positive height hhh, outputs all root bits, and restores all r+1r+1r+1 initially arbitrary dimension-ddd ring registers. Its simultaneous bounds are

S≤Cs(rℓ+h(L(k)+log⁡2m+log⁡2r)+s),S\le C_s\bigl(r\ell+h(L(k)+\log_2m+\log_2r)+s\bigr),S≤Cs​(rℓ+h(L(k)+log2​m+log2​r)+s), T≤Ct2Crhlog⁡2r(Bk)h(2rℓ+k+d+2s)q.T\le C_t2^{C_r h\log_2r}(Bk)^h(2^{r\ell}+k+d+2^s)^q.T≤Ct​2Cr​hlog2​r(Bk)h(2rℓ+k+d+2s)q.

All positive constants and the integer degree qqq precede the varying arity, frames, trees and register contents. There is no upper bound on rrr relative to ℓ\ellℓ. This companion theorem extends the binary framework and supplies the higher-arity bound used in blocking.

Differences from the source:

  • encoding: asymptotic algorithmic resource bounds → a finite binary machine and positive uniform constants, with fixed polynomial degrees where needed; reason: expands O/poly using actual transitions and visited work cells; constants precede varying inputs and tapes
  • encoding: uniform access algorithms and common space s in the section setting → a parameterized frame, two fixed binary printers, a common bound A, and auxiliary-parameter length at most 2^(A s); reason: makes Remark 3.8 access and input-size conventions explicit; constants depend on the fixed access procedures, never on their inputs
  • encoding: r-ary trees and r+1 registers of dimension d → Fin r child positions, lexicographic full tables and binary serialization of exactly r+1 original-dimension registers; reason: encodes the same data, extends the published binary preorder, and charges any initialized zero padding to local workspace
  • encoding: log k in bookkeeping → L(k)=1+log₂(k+1); reason: the additive constant is absorbed by log m and log r for m>1 and r≥2
Preamble
import Definitions.Def_CatalyticTreeEval_Main_Supplement_HigherFramework

set_option autoImplicit false
Formal statement
namespace CatalyticTreeEval.Main.Supplement

theorem theorem_4_11 (F : List Bool → Frame) (domain : Set (List Bool))
    (MV MD : Machine) (A : ℕ) (hA : 0 < A)
    (h : ∀ a ∈ domain, (F a).valid ∧ a.length ≤ 2 ^ (A * (F a).s) ∧ (F a).access a MV MD A) :
    ∃ (M : Machine) (Cs Ct Cr B : ℝ) (q : ℕ),
      0 < Cs ∧ 0 < Ct ∧ 0 < Cr ∧ 0 < B ∧ 0 < q ∧
      ∀ a ∈ domain, ∀ h : ℕ, 0 < h →
      ∀ (I : ArityTree (F a).r (F a).l h)
        (R : Registers (F a).r (F a).m (F a).d),
      RegisterComputes M (listCode ((F a).params a) ++ I.encode) (bits (F a).l I.value.val) R
        (Cs * ((F a).r * (F a).l +
          h * (lg (F a).k + log₂ (F a).m + log₂ (F a).r) + (F a).s))
        (Ct * exp₂ (Cr * h * log₂ (F a).r) * (B * (F a).k) ^ h *
          ((2 : ℝ) ^ ((F a).r * (F a).l) + (F a).k + (F a).d + (2 : ℝ) ^ (F a).s) ^ q) := by sorry

end CatalyticTreeEval.Main.Supplement
Source
Alexandra Henzinger, Edward Pyne, Seyoon Ragavan, Catalytic Tree Evaluation from Matching Vectors, STOC 2027 submission #533 (October 2026), canonical source.pdf SHA-256 17c3a71f0fb5afe2456558477e16fbac4200191e6e68f0facf212e24cb4ef238, Theorem 4.11, printed p. 28 (PDF p. 30)

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