Repository navigation
Conversation
Give a second complete width-four projector pool at the even matching maximum. Pair-product traces distinguish its action from the existing maximizer, and finite enumeration checks every lift and starting vector.
This was referenced Nov 18, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
This scoped construction gives a second complete binary projector pool with pattern dimension
k=3, matching numberν=1and irreducible joint-image dimensionn=4. Two unordered pairs of its projectors have a trace-one product. Four pairs do in the published maximizer. The trace count is preserved by simultaneous conjugation, so the two actions are inequivalent even after relabelling the projectors.The new construction was produced by Nick Petalas. Credit aliromman for the even maximum and comparison pool in #7707 and the matching theorem in #7692. Advance claim.
Evidence
The written proof finds the complete pool by its two cross rectangles and shows irreducibility by a directed walk through the four basis images. Both verifier modes scan all 120 rank-one idempotents at width four and find four lifts in the seven nonzero patterns. All 15 nonzero starting vectors reach all four basis images. Both modes print
PASS_EVEN_ONE_EDGE_MAXIMIZERand took 0.01 seconds each locally.Run from the repository root with CPython 3.12.15 and its standard library. Local verification ran on macOS arm64.
The upper bound is cited from the pinned matching result. The verifier checks the new witness. Predecessor replays, selected cold replay, Lean and Docker were unrun. Independent human and formal proof-kernel review are unclaimed. The native word, dirty-state restoration and recursive invoice remain open.
Open next questions
(3,1,4). Enumerate complete pools containing a fixed unit at width four before considering wider embeddings.Related open work
n=2k−1. Atk=3, its maximum uses two edges.(4,0,5)maximizer. The new alphabet has four projectors.CI notes
This PR carries the shared CI-doc and replay-gap sync used by other open PRs.
The six local current-status tests and the registered historical replay-gap check passed with CPython 3.12.15. All three hosted jobs passed at
6bd8a384d64b6da2334bc94f6a97cdd74d710780. The jobs wereplanand the registered replays forresearch/even-one-edge-maximizerandresearch/historical-selected-replay-gap. Run 44400399401Before the sync, the canonical base
e54523517d0f03e58352e647c403a5af3e525a57failedtest_current_ci_section_matches_selected_recordlocally because the CI guide namesmake frontier6030-verifywhile the selected record requiresmake frontier7000-verify. The same planning failure skipped hosted replay on #7740 and #7711.Merge danger
Two-way. The construction adds five files under
research/even-one-edge-maximizer/. The CI sync is a separate commit.