Skip to content

Classify constant insertions after stage-2 E_22 and the stage-2 leaf (no κ) - #290

Draft
unpaidmoon wants to merge 1 commit into
cursor/upstream-e545235-base-cbddfrom
cursor/stage2-e22-leaf-gaps-cbdd
Draft

unpaidmoon wants to merge 1 commit into
cursor/upstream-e545235-base-cbddfrom
cursor/stage2-e22-leaf-gaps-cbdd

Conversation

@unpaidmoon

Copy link
Copy Markdown
Owner

Change and mathematical scope

Scoped proof, no κ change. In the nine diagonal gates of PR CrocSwap#7027, constant matrices over F_2 placed in the gap after stage-2 E_22 and the gap after the stage-2 leaf gate make the coefficient of e equal the pure product P for exactly 2^40 pairs. Every diagonal entry of that coefficient lies in {0, P_ii}, and each P_ii differs from 1, so the coefficient is never I.

The ring is F_2[x, y, z][e]/(e^2) with e central. The word ends at the stage-2 leaf, so the factor after the second insertion is the empty product I. The middle diagonal is diag(1, 1, 1, 1, 1, 1, 1+z, 1+z). Matched off-diagonal positions are the pairs inside {6, 7} and inside {0, 1, 2, 3, 4, 5} (32 positions). The other 24 off-diagonal positions are zero in both matrices. The eight diagonal bits of R are free, with S_ii = 1 + R_ii.

This is not a listing of the 2^40 matrices, not a caller, and not a change to the selected frontier 363005817/100000000000. The selected record remains the PR CrocSwap#7000 / CrocSwap#7001 admission. current-priorities.md on this base still prints the older decimal 0.00310045; this package uses the selected record and does not edit it.

Task and overlap

Base: e54523517d0f03e58352e647c403a5af3e525a57. Priority revision 8. Card joint-query-controller, scope changed: the card asks whether a changed query demand removes real work, and this result is the second open question of PR CrocSwap#7287, head 0f1fc97b0fb02dbe8debb8ed1f480421d54027e2.

Session r281 took item 4 of the open-handoff list ordered newest package first, then README question order, among unpaidmoon open next questions not already delivered:

  1. Record that the V02 public replay does not bind the 56 assumptions CrocSwap/integer-mult-bounds#7288 Q1, transcript binding of the 56 assumption values.
  2. Record that the V02 public replay does not bind the 56 assumptions CrocSwap/integer-mult-bounds#7288 Q2, whether the two hashed pyconfig.h rows define the seven named macros.
  3. Classify constant insertions after stage-1 E_22 and stage-2 E_11. CrocSwap/integer-mult-bounds#7287 Q1, degree-at-most-1 solutions in the stage-1 E_22 / stage-2 E_11 gaps.
  4. Classify constant insertions after stage-1 E_22 and stage-2 E_11. CrocSwap/integer-mult-bounds#7287 Q2, constant F_2 pairs after stage-2 E_22 and after the stage-2 leaf gate.

Nearest records: PR CrocSwap#7287 classifies the earlier pair and leaves this one open. PR CrocSwap#7215 classifies the pair after stage-0 E_22 and stage-1 E_11. PR CrocSwap#7103 and PR CrocSwap#7191 classify the adjacent stage-0 pair and its polynomial entries. PR CrocSwap#7039 is the one-insertion diagonal obstruction, which still applies. GitHub search of open and closed PR titles and bodies found no other statement of these two slots. The count 2^40 and the level sets differ from CrocSwap#7287: (1, 4) is blocked there and free here, and (1, 6) is free there and blocked here.

Validation

Python 3.12, standard library only. From the repository root:

python3 -B scripts/research_package.py --package research/stage2-e22-leaf-gaps

Both the ordinary run and the python -O run print PASS_STAGE2_E22_LEAF_GAPS. The checker raises SystemExit, so optimized mode does not drop the checks. It rebuilds the nine diagonals, checks the closed form of P, the middle diagonal, the empty suffix I, the scale identity at all 64 positions, the direct product of all 64 matrix units in each gap, and all four constant bit pairs at every position. Witnesses with both matrices nonzero are multiplied directly. A blocked bit, an unequal matched bit, and a mixed pair are rejected.

Evidence class: a proof of the entry formula and of the constant solution set, with each scalar equation solved exhaustively on a domain of four bit pairs. It is not a heuristic search and not an enumeration of the 2^40 matrices. Not run: a replay of PR CrocSwap#7287 itself, a polynomial-entry classification, a caller or native bill, and any selected-frontier replay.

Credit

The paired-star contract is PR CrocSwap#5747 (Ali Romman), including PR CrocSwap#5673, PR CrocSwap#5631, and PR CrocSwap#5617. The nine-gate word is PR CrocSwap#7027. The one-insertion theorem is PR CrocSwap#7039. The stage-0 pair and its polynomial entries are PR CrocSwap#7103 and PR CrocSwap#7191. The stage-0 E_22 / stage-1 E_11 pair is PR CrocSwap#7215. The question answered here is the second open question of PR CrocSwap#7287.

Prepared by unpaidmoon with substantial Cursor Grok assistance. Independent human review and a proof-assistant proof are unclaimed. Apache-2.0.

Open next questions

  • Do degree-at-most-1 entries in these same two gaps give solutions of C = P beyond the 2^40 constant pairs? At (0, 6), R = 1+z and S = 1 solves that one blocked entry. Keep a legal constant diagonal and legal matched bits, substitute that pair, and multiply. PR Classify constant insertions after stage-1 E_22 and stage-2 E_11. CrocSwap/integer-mult-bounds#7287 leaves the same style of question for its own earlier slots; this package does not take those slots.
  • Which constant F_2 pairs make C = P in the gap after stage-2 E_11 and the gap after stage-2 E_22? The single gate between those slots is stage-2 E_22. This package classifies the following leaf instead.

Workflow feedback (optional)

The package generator and the registered replay hook were enough. No separate session report.

Open in Web Open in Cursor 

In the nine-gate diagonal word, constant F_2 matrices in the gap after stage-2 E_22 and the gap after the stage-2 leaf make the coefficient of e equal the pure product for exactly 2^40 pairs, and that coefficient is never I. No change to the selected frontier.

Co-authored-by: unpaidmoon <unpaidmoon@users.noreply.github.com>
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.

2 participants