Skip to content

Classify all 386 complete width-four one-matching projector actions - #7754

Open
aliromman wants to merge 1 commit into
CrocSwap:mainfrom
aliromman:codex/consecutive-result-069
Open

aliromman wants to merge 1 commit into
CrocSwap:mainfrom
aliromman:codex/consecutive-result-069

Conversation

@aliromman

@aliromman aliromman commented Nov 18, 2026 •

Copy link
Copy Markdown
Contributor

The width-four binary projector problem has exactly 386 complete action classes, including 40 irreducible classes. This finite classification answers the width-four part of Nick Petalas's open question in PR #7749. No new kappa or physical supplier is claimed.

Read the coverage and conjugacy proof, complete classification certificate, factor-pair census, separate matrix path, replay entry point, source pins and credits.

The exact class consists of three-dimensional off-diagonal pattern spaces in ambient F2^4, with matrix-unit graph matching number one, full joint image and the complete pool of rank-one idempotents with nonzero pattern. Equivalence means simultaneous GL(4,2) conjugacy of the unordered complete pools. Reducible actions are included. Canonical transformed factors can have zero off-pattern; the certificate therefore also gives an original qualifying representative for every class.

Pool members Reducible action classes Irreducible action classes
4 125 22
5 148 12
6 59 6
7 12 0
10 2 0

A coordinate permutation puts a present unit at E_01. The first path checks all 698,027 two-spaces of the eleven-dimensional quotient and finds 2,070 qualifying anchored spaces. A separate path scans all 65,536 binary matrices, identifies the 120 rank-one idempotents by their rows and products, then enumerates spans of realized patterns. The complete dictionaries agree.

Every anchored space has an explicit conjugacy witness. For each of the 386 class representatives, a separate matrix calculation checks all 20,160 invertible matrices, both inverse products, the full-group minimum and the stabilizer. All minima differ. Testing all 67 subspaces per representative gives the irreducible count. The source verifier runs unchanged; the two known pools occur literally as classes C347 and C384.

Python 3.12, standard library only:

python3.12 -B scripts/research_package.py --package research/width-four-projector-classification

Both registered commands print:

PASS_WIDTH_FOUR_PROJECTOR_CLASSIFICATION
698027 planes; 2070 anchored spaces; 386 action classes; 40 irreducible classes
120 projectors; 20160 direct matrix conjugacies per class; 67 invariant subspaces per class
Source classes: {"PR7707": "C384", "PR7749": "C347"}

Normal and optimized registered replay, syntax/indentation, committed bytes/source history/modes, export and sealed read-only Git-free serial replay pass. Matching-two and pattern-dimension-two controls are rejected. 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.

Nick Petalas supplied the second attaining action and classification handoff in the source proof, head 6bd8a384d64b6da2334bc94f6a97cdd74d710780. Ali Romman's PR #7707, head 57e24923b609d2588b83321f530fa2a1e75ed364, supplies the even attaining family. His PR #7692, head c2870a38841f515db7203a0761cd77f014f7fb7a, supplies the matching bound. All nine retained files and their predecessor credit remain intact.

This does not classify larger ambient embeddings. Reducible width-four actions are not global maximizers under the general matching bound. Useful integral exponent relations, paid words, native realization, dirty restoration and the complete recursive invoice remain open. Historical full suites, selected cold replay, Lean and Docker were not run. The selected frontier and all five framework assumptions remain unchanged.

Base: canonical main e54523517d0f03e58352e647c403a5af3e525a57, priority revision 8. No unmerged source ancestry is asserted. Prepared by Ali Romman with substantial OpenAI Codex assistance in research, implementation, verification and writing. The two computational paths share the mathematical definitions and fixed width; no independent human or formal-kernel acceptance is claimed. Apache-2.0. Canonical claim.

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