Repository navigation
Conditional κ = 0.00165861362787227: three entrance-promotion groups on the PR659 no-triangle word - #664
Conditional κ = 0.00165861362787227: three entrance-promotion groups on the PR659 no-triangle word#664jeremykothe2 wants to merge 1 commit into
Conversation
…s on the PR659 no-triangle word
|
Thank you @jeremykothe2. This contribution is now retained on main in the scoped 100-PR review and contributor credits. Contribution: Three entrance-promotion groups on the no-triangle word. Review and limits: The scalar program stays fixed but all changed entrance/frame charges and owner packing are recomputed. Full promotion's exclusion is a restricted packing result. The complete original source, notices and relevant execution evidence are retained through the pinned source archive. The review distinguishes fresh local runs, source-bound historical Linux CI and prior exact-head review; it does not claim 100 fresh full mathematical audits. Closing this historical PR as reconciled. This preserves parallel and cumulative credit without replacing the selected frontier or promoting priced research targets to completed constructions. Reviewed head: |
Summary
Conditional κ = 0.00165861362787227 (conservative 0.00165861362498948). This is +2.42×10⁻⁶ over PR659 (0.001656193765416678). The complex side binds; PR589's bit supplier gives 0.0017334.
Change: entrance promotions on PR659's word. An entering helper group whose entrance frame is E4 starts at E5 instead, with its first gates raised so that the entrance conditions (zero target response, monotone frames) still hold. This is the time-reverse dual of deepening an end. The scalar program is byte-identical to PR659's word (sha a9de804f); only frame annotations and entrance frames change. Each promoted group gives Δ{1:+8, 2:−8, 3:−4, 4:+4} and width −4 per call.
Three groups are promoted. Promoting all eleven groups loses: narrowing 5→4 leaves odd-width gaps in the rank-32 bins that no owner assignment can fill. An exact packing relaxation bounds full promotion below PR659. Three groups pack cleanly.
Construction:
construction/promo/addsbuild_promotion.py, a generic driverreproduce_promo.py --k, the k=3 recipe and local pins, andpromo_controls.py. The controls reject 10/10 mutations at pinned sites: wrong entrance frame, a missing raised gate per group, over-promotion, and an input-pin control. The recipe carries 72 additional price-neutral frame adjustments from its CP-SAT search, documented inPROOF-NOTE-PROMO.md; a minimal deterministic recipe gives the identical profile. All verifier code is byte-identical to PR659's reviewed set.Independent review: SOUND. The reviewer confirmed the scalar program is unchanged, re-ran the exact replay (0 bad rows, dyadic, all orientations), (N) 21/21, the payload bound and audit_local, and found every mutant rejected by both the native and the independent audit. Its own local chains are byte-identical apart from timing fields, and its own strict runs reproduce this κ. It also ran the packaged clean rebuild.
Verification
From a fresh clone of commit fed4c13, with no reused artifacts (Windows, CPython 3.11.11):
verify_manifest.py: PASS, 111 filesreproduce.py, thenreproduce_notri.pyandreproduce_np.py: PR659's word rebuilt; PASS layer-end-controlsreproduce_promo.py --k 3: promoted word rebuilt, promo controls rejected, PASS wrappers, PASS phasesaudit_local.py: PASS_INDEPENDENT_EXACT_FORWARD_AND_DUAL_ENDPOINT_FRAMES_ENTRANCES_CONESverify_global.py: PASS invoice, assembly, admission-controls and bit-supplier-controls; PASS_GAUGE44_NATIVE_EXACT_GLOBAL_ASSEMBLY 0.00165861362787227The rebuilt locals equal the strict locals except two
secondstiming fields. The exact commands are inresearch/gauge44-promo/README.md.Scope
Conditional on PR645's premises and on PR589's bit-supplier assumptions. No unconditional theorem or Lean certification is claimed.
Credits
Prepared with Claude Code (Anthropic).