Repository navigation
Independent exact forward-basis audit of PR1611 helper450 - #1710
Merged
Merged
Conversation
Contributor
|
Batch review at Merged to Reviewed scope: Independent forward dyadic replay checks all 712 entries of helper450 output row and its dirt column in both orientations, corroborating PR1611. Source producers are shared; event-to-Lean correspondence and physical hosting remain unproved by this scalar audit. The pinned scoped command passed; the batch ledger contains its command, input hashes, environment and log. This incorporates the research/tooling contribution and does not change the selected κ or discharge the inherited multiplication-theorem assumptions. Batch review, attribution and evidence. Thank you for the contribution. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Change and mathematical scope
PR1611 certifies helper450's scalar restoration using sparse rational row/column propagation. This independently corroborates that scalar premise by pushing every input basis vector forward with a separately written exact dyadic implementation. No discrepancy was found. Only
research/k8-hpe450-forward-audit/is added: eight files, including the checker, deterministic receipt, source pins and review/validation records.Both pinned local orientations have 24,767 events and 19,693 scalar operations. All 1,424 basis executions confirm the unit helper output row, its isolated dirt column and zero coefficients from all eight arbitrary private inputs. Six malformed programs reject with explicit exact coefficient witnesses. A dirt-leaking read preserves the row but leaks 1/2 to L230 forward and 1 to L27 backward. This is finite corroboration, not a new identity, Lean theorem, supplier admission or kappa.
The retained conditional C7 conclusion still needs exact event recipes and JSON-to-Lean correspondence. COPY/CLEAR lifetimes, other recipes, hosting/footprints, all-size compilation, precision/tape, complete costs and analytic transfer remain open. Selected sources/frontier and predecessor packages are unchanged.
Task and overlap
Base current main
0b1cccf86e2f5132ed286a26ef0a681a495740ea; contributiona2078691245ff8d2825def7048680759b695cb49. Taskformal-interface-lemma, scope keye3ffea67887eeb4c396c, nonce9c92a140d6ff014ccc0bdd30a86e9981, revision 3. Scope refined to the scalar premise of one existing theorem-to-emission binding; exact card and stop rule are in TASK.json. Claim.156cc788b5a8122b19686f015e5654b117e4389d: construction/binding predecessor; the new algorithm obtains the row by forward basis execution rather than reverse covector propagation. Its published verifier separately passes all 28 controls. No pending checker code is imported.0d8d2c3a96346756f02563b401a042ab97198e20: retained complex operator semantics, not repeated.Final overlap at 07:25 UTC read 393 open PR bodies, including drafts, through an empty numbered page; the helper also searched 875 local records. Discussions1163/1166/1394 and the active board were read live. Full Discussion pagination was unavailable, the helper warns about its absent Discussion cache, and concurrent merges make numbered pagination non-atomic. No-match output is not a novelty certificate.
Validation
Registered replay passes on CPython3.12.14/macOS in 5.72s. Also passed: metadata/source pins, unchanged selected-record check, 18 repository onboarding tests, a read-only optimized source export without Git, source-drift and receipt-corruption rejection, and occupied-output preservation. Exact arithmetic is compared with Fraction over 4,624 operand pairs; a four-input dirty echo checks destructive COPY/CLEAR and inverse scales. Two bounded agent static reviews found no actionable issue. VALIDATION.json records actual commands/environment and separates negative exit codes from success.
No fresh Lean/Mathlib build, general serialization proof, physical hosting replay, full selected cold replay, hosted CI success or independent human review is claimed.
Credit
Hanson Sun/@bbMisk (PR1611); Jeremy Kothe (HPE/k8); DreamingOfClouds (wrappers/Lemmas A/G); icekylinx (compiler); Rohan Arun (packing); Josie Elliston and Sagar Thakrar (word/audit lineage); @commdata2338 (PR1415); Aaron Wang/navinor (distinct cold audit); Chafik Boukhalfa, Zhihao Chen and all retained realization contributors. New work by @ToikY with substantial OpenAI Codex assistance in research, implementation, verification and writing, including bounded subagents. Apache-2.0; original notices unchanged.
Open next questions
README records the smallest useful follow-ups: translate this one helper into an actual Lean event word with private COPY/CLEAR lifetimes, or trace it through actual T16 hosted invocations. Scalar agreement does not discharge either missing correspondence or physical footprint proof.