Skip to content

Prove sharp no-edge total-image bounds for every pattern dimension (no κ) - #7690

Open
aliromman wants to merge 1 commit into
CrocSwap:mainfrom
aliromman:codex/consecutive-result-030
Open

aliromman wants to merge 1 commit into
CrocSwap:mainfrom
aliromman:codex/consecutive-result-030

Conversation

@aliromman

@aliromman aliromman commented Nov 18, 2026 •

Copy link
Copy Markdown
Contributor

Result and scope

The sharp no-edge total-image bound is floor(3k/2) at every pattern dimension k. This holds for any binary rank-one idempotent collection whose patterns are nonzero and whose pattern span contains no matrix unit. Completeness and an irreducible image-basis subcollection are not required. If the action on a nonzero joint image is irreducible, the bound is floor((3k-1)/2).

Read the proof, exact receipt, verifier and scope record. The proof shows that all pattern dependencies of an independent image subcollection are generated by disjoint circuits of at least three members. A coordinate-column argument treats the full diagonal-relation space at once. The all-width proof uses no finite classification step.

At pattern dimension five, both the total and irreducible image maxima are seven. This closes the upper-bound question in PR #7281; its existing seven-image construction supplies prior sharpness evidence. The new theorem also extends the dimension-three/four support bounds and applies when the irreducible basis extraction needed by PR #7686 is unavailable. It is consistent with the extraction counterexamples in PR #7688.

No new kappa, paid BIT word, native operation or full invoice is supplied. The selected frontier and all five assumptions remain unchanged. Coordinate classification of all attaining dimension-five pools is still open. This completes the advance claim.

Evidence and reproduction

Normal and optimized CPython 3.12.16 replay passed on macOS arm64. Complete image-basis censuses at widths three and four cover 224036 collections with at most three members. All 95046 collections with nonzero unit-free pattern spans pass the full diagonal-relation certificate, including disjoint active supports and the entire relation code.

All 2825 width-three pattern spaces are enumerated. Of these, 1330 are unit-free, 336 have empty pools and 595 have irreducible nonempty pools. Every complete pool satisfies both applicable bounds. Complete sharp pools for 1<=k<=12 cover 8178 nonzero patterns. Larger odd, even and mixed circuits, five equality fixtures with overlapping-support controls, and the prior seven-image witness also pass. Matrix-unit, non-idempotent and zero-pattern controls violate the bound outside its hypotheses.

Metadata, committed bytes, modes, source ancestry, eight-file export and archive integrity pass. A read-only optimized export replay passes without Git or a usable PATH; all nine payload/manifest hashes and modes stay unchanged. The finite checks corroborate the written theorem and are not its all-width argument.

With CPython 3.12 and its standard library, from the repository root:

python3.12 -I -W error -B research/no-edge-total-image-bound/verify.py
python3.12 -I -W error -B -O research/no-edge-total-image-bound/verify.py
python3.12 -B scripts/research_package.py --package research/no-edge-total-image-bound
python3.12 -B scripts/check_research_commit.py --package research/no-edge-total-image-bound --check-source-history --check-modes

Expected markers:

PASS_NO_EDGE_TOTAL_IMAGE_BOUND
PASS_REGISTERED_RESEARCH_PACKAGE
PASS_RESEARCH_COMMIT_BYTES

Hosted CI did not certify this package. The plan job failed in the shared test_current_ci_section_matches_selected_record: the base guide names frontier6030 while the selected record requires make frontier7000-verify. The log reports 760 tests, one failure and 24 skipped; package replay was skipped. This package changes no shared CI file. Local replay remains the scientific evidence. Predecessor scientific replays, the selected cold replay, Lean and Docker were not run. Independent human and formal-kernel review are unclaimed.

Sources and credit

Prepared by @aliromman with substantial OpenAI Codex assistance. @CubeJerry supplied the intersecting diagonal-dependency method in PR #6838, exact proof. @unpaidmoon supplied the triple/support analysis and dimension-four bound in PR #7139, exact proof, and the seven-image construction in PR #7237 and PR #7281, exact source. Nick Petalas (@petalas) supplied the fiber lemma credited by those exact sources; a short proof is included here.

Ali Romman's PR #7686, exact proof, and PR #7688, exact proof, supply the immediate scope boundary. Their matrix/lift routines and subspace routine are reused with credit. Earlier bounds, examples and normal-form classifications are not new here. New work is Apache-2.0; no predecessor package is needed at runtime.

Base: canonical main e54523517d0f03e58352e647c403a5af3e525a57, priority revision 8. Card bit-new-register-model, scope changed to this scalar structural theorem; the native construction gate is unmet. Refreshed open/closed PR and Discussion overlap and live handoffs identify the bounded predecessors and no competing general-bound claim. Sparse local and unpublished work remain coverage limits. The next classification can start from two relation triples plus one free member, or a triple plus a quad, instead of an unbounded search for a larger image.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant