Repository navigation
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Scoped written proof with exact finite corroboration, produced by Nick Petalas. For heterogeneous diagonal Coxeter targets that are subsequences of one order of distinct simple generators, the exact formal price at rank cap
bisHere
d_scounts the target roles requesting generators. The common pricefhas nonnegative, nonincreasing increments andf(0)=0. The proof works over every field with arbitrary scalar couplings, free scalar invertible maps and any finite number of restored dirty roles. A cut on the cosets of the parabolic subgroup omittingsgives a separate rank demand for that letter. Batching the requesting roles in the common order attains all the demands.For the mixed A3 tuple
(s, st, tu, su, stu)at cap two andf(k)=k/(k+1), the exact price is 11/3. The pooled root-orbit bound is 10/3 and separate target calls cost 5. The proof answers the common-order part of zhaobenny's heterogeneous-target question in #6817, using ToikY's #6821 as the repeated-target precedent. Credit adamgall #1941 and RossRichardson #1966/#1989 for separate-letter cuts and DavidAriasC #2738 for Coxeter tariffs. Proof and checker, advance claim.Evidence
CPython 3.12.15 on macOS arm64. Both direct checks took 0.11 seconds locally and printed identical
PASS_COXETER_COMMON_ORDER_TARIFFresults. Each checked 10,512 full formal A3 endpoints, 873 paid-factor cut ranks and seven rejecting controls on the mixed example. The registered runner printedPASS_REGISTERED_RESEARCH_PACKAGE. The committed-byte, mode and source-history check printedPASS_RESEARCH_COMMIT_BYTESfor all five package files.Run from the repository root:
Limitations
The proof prices the declared formal group-algebra model. Native implementation, physical movement and scalar-map prices, precision, compilation and supplier admission remain open. The selected multiplication claim retains its five assumptions. Predecessor scientific replays, the selected cold replay, Lean and Docker were outside this check. Independent human and formal proof-kernel review are unclaimed.
Open next questions
(st,ts)at cap two, where individual rank cuts do not encode the chronology.Related open work
sttargets with independent monotone price tables and separate caps.CI notes
This PR carries the shared CI-doc and replay-gap sync used by other open PRs.
All three hosted checks passed at
112828b98bdb978b16776b6e1cd73453f9b1aab6. The plan job and the registered replays forcoxeter-common-order-tariffandhistorical-selected-replay-gapcompleted successfully. Run 44393751192.All six local navigation tests passed after the sync. The local historical-selected-replay-gap package, run with CPython 3.12.15 and
--install-deps, printedPASS_REGISTERED_RESEARCH_PACKAGE.On canonical main, the local navigation test
test_current_ci_section_matches_selected_recordfails because the CI guide namesmake frontier6030-verifyand the selected record requiresmake frontier7000-verify. The same assertion stopped the plan job in #7734, which reported one failure in 760 tests and 24 skips. Run 44389148736.Merge danger
Reversible. The proof adds five files in one research package. The separate CI commit updates five shared files.