Repository navigation
Scoped negative result: X4 target-forest floor holds for all dyadic parent coefficients (no new kappa) - #961
Merged
Mister-Meeseeks merged 1 commit intoOct 13, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The pinned PR799 X4 target-forest result (#932) establishes a 10,883-primitives-per-real-component optimum only for signed-unit parent coefficients. This research note extends that exact floor to every nonzero dyadic parent coefficient under the same independent binary-Horner lowering and leaf-before-parent forest schedule. No new kappa or supplier integration is claimed.
Two elementary tail bounds reduce all dyadics to 1,086 coefficients. Two separately implemented exact checkers cover 13,501,152 ordered coefficient/target-pair cases. 970 nonunit edges beat their original root, but every nonunit edge is strictly dominated by a root or signed-unit edge to the same parent. A forest exchange argument transfers the 10,883 floor; independent Prim/Kruskal and 12,544 cycle inequalities recheck the attaining signed-unit graph.
Reproduce offline, standard-library Python 3.10+:
The full checker passes with CPython 3.11.0 and 3.14.5; nine package tests pass. The original PR932 full scalar witness and selected-record verifier also pass. Selected-result tests pass; two existing symlink tests skip in the Windows sandbox. Full repository/cold-supplier suites and hosted CI were not run because no existing supplier or generator changes.
Base: 018418f; source/dependency pins, complete coefficient certificate, proof, local sprint claim (seed 31), controls and scope are in the new package. Credit Ross Richardson #932, ToikY #909, the restored-pair/forest predecessors and Jeremy Kothe/Chafik Boukhalfa/icekylinx supplier lineage. Substantial OpenAI Codex assistance, including a separate checker implementation and mathematical review, is disclosed. No independent human review or full multiplication-theorem formalization is asserted.
Multiple parents, helper-basis changes, arbitrary sequential shears, shared coefficient work and other lowerings remain outside scope. The selected frontier and upstream/ remain unchanged.