Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions reviews/lean_cycle2_20261005/agent7/BUILD_STATUS.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,16 @@
# BUILD_STATUS — L7 / MT-N1

**COMPILE_STATUS:** UNCOMPILED
**KERNEL_CHECKED:** NO
**PROVED_CLAIM:** NO (proposal only)

Frozen Math-#275 head: `7206526ed53f7539838711ee45257a13b4b07e99`
elan on box: 4.2.4
default Lean/lake toolchain: ABSENT (`elan` reports no active toolchain)
lake build of companion: NOT EXECUTED

No type/axiom/build log is attached because no successful (or failed) kernel
run of this companion was produced. Do not infer success from hosted #275 CI.

Blocker for integrator: configure toolchain, import companion on an authorized
branch, then run `lake build` / leanchecker (integration owner only).
Original file line number Diff line number Diff line change
@@ -0,0 +1,86 @@
/-
L7 / MT-N1 PROPOSAL — UNCOMPILED
Frozen Math-#275 head: 7206526ed53f7539838711ee45257a13b4b07e99
Source ask: main#229 comment 5995761095
Worker: Grok Bot agent 7 (non-Claude, nonauthor lane)
Status: UNCOMPILED — not kernel-checked; do NOT treat as accepted Lean.
Do not edit/rename the accepted theorem p02_lm009_moment40_family_r3.
This file is an isolated companion proposal under reviews/ only.
-/

import ResearchFormalCoreR1.MomentTail

namespace ResearchFormalCoreR1

open MeasureTheory
open scoped ENNReal

set_option autoImplicit false

variable {Ω : Type*} [MeasurableSpace Ω]

/-- Interval-only companion to `p02_lm009_moment40_family_r3` (MT-N1).

Same conclusion shape (one `C` fixed from `M,cW,cZ,ε`, cubic bound on
`0 < r ≤ r0`), but probability is required **only** for those `r` in the
interval — introduced as an explicit hypothesis inside each quantified case
of `hdata`, not as a global instance `[∀ r, IsProbabilityMeasure (μ r)]`.

Rationale (frozen #276 REVIEW.md §MT-N1): the accepted family theorem's type
demands `μ r` be a probability measure for every real `r`, while premises and
conclusion quantify only over `0 < r ≤ r0`. That extra instance is stronger
than needed; it is not silently relaxed here by editing the frozen theorem.

This declaration is separately named. It does not rename, edit, or replace
`p02_lm009_moment40_family_r3`. Unweighted / finite-measure (MT-N2) work is
out of L7 scope.
-/
theorem p02_lm009_moment40_family_r3_interval
(μ : ℝ → Measure Ω)
(W R : ℝ → Ω → ℝ) (r0 ε M cW cZ : ℝ)
(hr0 : 0 < r0 ∧ r0 ≤ 1) (hε : 0 < ε)
(hM : 0 ≤ M) (hcW : 0 ≤ cW) (hcZ : 0 < cZ)
(hdata : ∀ r, 0 < r → r ≤ r0 →
IsProbabilityMeasure (μ r) ∧
MemLp (W r) 2 (μ r) ∧ (0 ≤ᵐ[μ r] W r) ∧ Measurable (R r) ∧
Integrable (fun x => R r x ^ 40) (μ r) ∧
(∫ x, R r x ^ 40 ∂(μ r)) ≤ M ∧
(∫ x, W r x ^ 2 ∂(μ r)) ≤ cW * r ^ 4 ∧
cZ * r ^ 2 ≤ ∫ x, W r x ∂(μ r)) :
∃ C : ℝ, 0 ≤ C ∧ ∀ r, 0 < r → r ≤ r0 →
IsProbabilityMeasure (weightedLaw (μ r) (W r)) ∧
(weightedLaw (μ r) (W r) (badJetEvent (R r) r ε)).toReal ≤ C * r ^ 3 := by
refine ⟨(Real.sqrt cW / cZ) * Real.sqrt (M / ε ^ 8), by positivity, ?_⟩
intro r hr hrr0
obtain ⟨hProb, hW, hW0, hR, hRi, hMb, hsecond, hlower⟩ := hdata r hr hrr0
-- Instance for this r only (inside the quantified case), not for all ℝ.
letI : IsProbabilityMeasure (μ r) := hProb
exact p02_lm009_moment40_weighted_r3 (μ r) (W r) (R r) hW hW0 hR
r ε M cW cZ hr (hrr0.trans hr0.2) hε hM hcW hcZ hRi hMb hsecond hlower

/-- Optional bridge: the accepted global-instance family theorem follows from
the interval companion once probability is supplied for every real `r`.
Kept as a named corollary for the L7 checklist; not a rewrite of the frozen
module. -/
theorem p02_lm009_moment40_family_r3_of_interval
(μ : ℝ → Measure Ω) [∀ r, IsProbabilityMeasure (μ r)]
(W R : ℝ → Ω → ℝ) (r0 ε M cW cZ : ℝ)
(hr0 : 0 < r0 ∧ r0 ≤ 1) (hε : 0 < ε)
(hM : 0 ≤ M) (hcW : 0 ≤ cW) (hcZ : 0 < cZ)
(hdata : ∀ r, 0 < r → r ≤ r0 →
MemLp (W r) 2 (μ r) ∧ (0 ≤ᵐ[μ r] W r) ∧ Measurable (R r) ∧
Integrable (fun x => R r x ^ 40) (μ r) ∧
(∫ x, R r x ^ 40 ∂(μ r)) ≤ M ∧
(∫ x, W r x ^ 2 ∂(μ r)) ≤ cW * r ^ 4 ∧
cZ * r ^ 2 ≤ ∫ x, W r x ∂(μ r)) :
∃ C : ℝ, 0 ≤ C ∧ ∀ r, 0 < r → r ≤ r0 →
IsProbabilityMeasure (weightedLaw (μ r) (W r)) ∧
(weightedLaw (μ r) (W r) (badJetEvent (R r) r ε)).toReal ≤ C * r ^ 3 :=
p02_lm009_moment40_family_r3_interval μ W R r0 ε M cW cZ hr0 hε hM hcW hcZ
(fun r hr hrr0 =>
⟨inferInstance, (hdata r hr hrr0).1, (hdata r hr hrr0).2.1,
(hdata r hr hrr0).2.2.1, (hdata r hr hrr0).2.2.2.1,
(hdata r hr hrr0).2.2.2.2.1, (hdata r hr hrr0).2.2.2.2.2.1,
(hdata r hr hrr0).2.2.2.2.2.2⟩)

end ResearchFormalCoreR1
97 changes: 97 additions & 0 deletions reviews/lean_cycle2_20261005/agent7/NOTES_MT_N1.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,97 @@
# L7 / MT-N1 — interval-only family formulation

**Status:** UNCOMPILED proposal (`.lean.txt`). Scientific effect **NONE**.
**Worker:** Grok Bot agent 7 (Grok Bot support agent; non-Claude, nonauthor lane)
**Pickup:** main#229 [5996231433](https://github.com/d6g8k5htny-coder/main/issues/229#issuecomment-5996231433)
**Readback:** [5996234080](https://github.com/d6g8k5htny-coder/main/issues/229#issuecomment-5996234080)
**Ask:** [5995761095](https://github.com/d6g8k5htny-coder/main/issues/229#issuecomment-5995761095)
**CoS route:** [5996179903](https://github.com/d6g8k5htny-coder/main/issues/229#issuecomment-5996179903)
**Frozen head used:** `7206526ed53f7539838711ee45257a13b4b07e99` (Math-#275 live head)
**Alignment note:** #276 REVIEW.md §MT-N1 at
`reviews/moment_tail_alignment_20261005/REVIEW.md`
(REVIEW commit `0cb17c1b836651144486d55788eef8d7cf532081`,
record tip `3718fc1564c537971636a3f4091224964e20f9f9`)

## What changed (proposal only)

| Item | Accepted (frozen) | L7 companion |
|---|---|---|
| Name | `p02_lm009_moment40_family_r3` | `p02_lm009_moment40_family_r3_interval` (new) |
| Probability instance | `[∀ r, IsProbabilityMeasure (μ r)]` for **all** `ℝ` | `IsProbabilityMeasure (μ r)` **inside** `hdata` for each `0 < r ≤ r0` only |
| Fixed outside `r` | `M, cW, cZ, ε, r0`, witness `C` | **exact same** |
| Conclusion | `∃ C, ∀ r ∈ (0,r0], weighted cubic ≤ C r³` | **exact same shape** |
| Proof body | call `p02_lm009_moment40_weighted_r3` per `r` | same call, with `letI` from the per-`r` hypothesis |

No edit/rename of the accepted theorem. No unweighted/finite-measure theorems
(MT-N2 is agent6). No formal-source / manifest / workflow edits on this lane.

## Why this is a real relaxation (not wording)

The frozen family type carries a section/instance binder

`[∀ (r : Real), IsProbabilityMeasure (μ r)]`

that is elaborated for every real `r`, including values outside `(0, r0]`.
The bound premises (`hdata`) and the conclusion quantify only over
`0 < r ≤ r0`. No step of the proof uses probability for an out-of-interval
`r`. Requiring a global instance therefore excludes families that are
probability measures on the working interval but arbitrary (e.g. zero)
elsewhere — a strictly stronger hypothesis than the stated analytic scope in
`MOMENT_TAIL.md` / MT-N1.

The L7 companion moves the probability requirement into `hdata`'s quantified
case. `M, cW, cZ, ε` and the witness `C` remain parameters **outside** `r`
(no per-`r` constant; no new uniformity premise).

## Outside-interval zero-measure test (why the relaxation is real)

Fix any `r0 ∈ (0,1]`, and define a family `μ : ℝ → Measure Ω` by:

- for `0 < r ≤ r0`: `μ r` is any probability measure satisfying the usual
`hdata` package (e.g. a Dirac probability used in frozen finite models);
- for every other real `r` (including `r ≤ 0` and `r > r0`): `μ r = 0`
(the zero measure).

**Under the accepted theorem type:** the instance
`[∀ r, IsProbabilityMeasure (μ r)]` **fails**, because the zero measure is
not a probability measure (`μ(Ω) = 0 ≠ 1`). The family cannot even be fed to
`p02_lm009_moment40_family_r3`, despite agreeing with every in-interval
hypothesis and conclusion obligation.

**Under the L7 companion:** `hdata` only asks for `IsProbabilityMeasure (μ r)`
when `0 < r ≤ r0`. Outside that interval the zero measure is allowed; the
existential `C` and the in-interval cubic bound are unchanged. Therefore the
companion admits a model the frozen type rejects — a real relaxation, not a
rewording of the same type.

(This test is analytic / type-theoretic. It is **not** a Lean kernel run of
the companion. Zero measure outside the interval is also the explicit L7 ask
in 5995761095.)

## Compile / toolchain honesty

| Probe | Result |
|---|---|
| `elan` | 4.2.4 present on box |
| default toolchain | **NONE configured** (`lean`/`lake` error: no default toolchain) |
| Hosted #275 Lean (for context only) | 4.34.1 / mathlib `d13f23b7…` on frozen CI — **not** a kernel receipt for this companion |
| Local kernel build of companion | **NOT RUN** |
| Unrelated green CI | Must **not** be cited as kernel evidence for this proposal |

**Verdict for integrator:** treat as **UNCOMPILED** `.lean.txt`. Register/audit
only after source review and a real kernel execution on an authorized formal
path. Proof-containing text ≠ accepted Lean.

## Out of scope (respected)

- formal-source / manifest / workflow edits
- merges; flag flips (`lemma_closed`, `prizes_solved`, `discharges_OBL_H5_JETMOD`,
`certified_C_H`, freeze, `inventable_attempt_accepted`)
- MT-N2 finite-measure unweighted companion (agent6)
- SoT / scientific-status invention; OBL stays OPEN

## Files in this directory

- `MomentTail_family_interval_MT_N1.lean.txt` — proposed companion (UNCOMPILED)
- `NOTES_MT_N1.md` — this file
- `BUILD_STATUS.md` — compact compile status receipt
Loading