Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 18 additions & 1 deletion docs/ci-verification.md
Original file line number Diff line number Diff line change
@@ -1,8 +1,25 @@
# CI verification

## Current selected-family replay: #6030
## Current selected-family replay: #7000 with #7001

The [selected record](../certificates/selected-result.json) binds conditional
κ = 0.00363005817 to the [application review](research/pr-intake-evidence/pr7000-application/PROOF-REVIEW.md),
[validation receipts](research/pr-intake-evidence/pr7000-application/VALIDATION.json),
and [explicit source closure](../certificates/frontier7000-source.json).
`make frontier7000-verify` follows that exact-source selected replay. The review
reuses eleven authenticated preceding executions and records seven fresh serial
application, scalar and complete native-catalogue executions. The earlier
monolithic timeout is retained separately and is not completed coverage.
Consult the receipts for exact commands, source heads and limits.

The five inherited framework assumptions remain. No complete uniform word,
literal multiplier execution, independent human proof or formal-kernel proof
is asserted. Routine research submissions do not need to rerun this
selected-family replay. The sections below preserve historical checkpoints.

## Historical selected-family replay: #6030

The [historical checkpoint](../certificates/checkpoints/frontier6030.json) binds conditional
κ = 0.003125804 to the [proof review](research/pr-intake-evidence/pr6030-audit/PROOF-REVIEW.md),
[validation receipts](research/pr-intake-evidence/pr6030-audit/VALIDATION.json),
and [explicit source closure](../certificates/frontier6030-source.json).
Expand Down
140 changes: 140 additions & 0 deletions research/coxeter-common-order-tariff/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,140 @@
# Exact rank-capped tariffs for Coxeter targets with a common generator order

This scoped proof gives the exact formal price of a heterogeneous diagonal
target whose entries are subsequences of one order of distinct simple
generators. It applies to every field and every finite-rank Coxeter system,
including infinite groups. The number of restored dirty roles is arbitrary
and finite.

## Model and result

Let `(W,S)` be a Coxeter system and work in `M_r(K[W])` over a field `K`.
A paid factor for `s in S` is `I + ([s]-[e])A`, where `A` is a scalar matrix
of rank between 1 and `b`. The model allows non-idempotent and singular
couplings. Scalar invertible maps interleaved with the paid factors are free.
All initial roles are arbitrary. The required endpoint is the full formal
diagonal with target `[w_j]` on role `j` and `[e]` on each spectator.

Fix an ordered list of distinct simple generators. Assume each `w_j` is the
product of a subsequence of this list, in the given order. Let `d_s` count
the roles whose subsequence contains `s`. Each call of rank `k` costs
`f(k)`, where `f(0)=0` and the increments of `f` are nonnegative and
nonincreasing. The exact price is

```text
sum over s in S of (floor(d_s / b) f(b) + f(d_s mod b)).
```

### Lower bound

For a given generator `s`, put `H = <S minus {s}>`. The generator `s`
lies outside `H`. In the real geometric representation, every other
simple reflection changes only its own simple-root coordinate and hence
preserves the `s` coordinate. The reflection `s` sends its simple root
to its negative. This group fact precedes the choice of coefficient
field `K`. The real geometric representation is the one used in
[David Arias's Coxeter tariff](../coxeter-root-tariff/PROOF.md).

Use the left action of `W` on the basis of left cosets `gH` over `K`.
Tensor the action with the role space `K^r`. Separate the copy of `K^r`
over `H` from all other cosets, and let `q(X)` be the rank of the block
from that copy to the rest. With the base coset indexed by 0 and
the rest by 1, the crossing block of `XY` is
`X_10 Y_00 + X_11 Y_10`. Subadditivity of rank gives
`q(XY) <= q(X) + q(Y)`. This holds on the algebraic direct sum for
infinite `W`, since the block's domain is finite-dimensional.

A paid factor labelled `t != s` has zero crossing rank because `t`
fixes the coset `H`. Scalar maps also have zero crossing rank. A paid
`s` factor sends its crossing block entirely to `sH` with matrix `A`.
Its crossing rank is exactly `rank(A)`.

A target that omits `s` fixes `H`. A target that uses it has the form
`a s c` with `a,c in H`. It moves `H`, since `a s c in H` would imply
`s in H`. The crossing images of the `d_s` target columns belong to
distinct role coordinates, so the target crossing rank is `d_s`. Hence
the sum of the ranks of the paid `s` factors is at least `d_s`.

Apply concave packing separately to each generator. Given a fixed rank
mass, transferring a unit from a smaller nonempty part to a larger part
below the cap does not increase cost, by the
nonincreasing increments. This leaves full cap-sized parts and at most
one remainder. The resulting packed price is nondecreasing in the mass,
so mass at least `d_s` costs at least the displayed term. Sum over the
generators. [Benny Zhao's concave packing bound](../coxeter-capped-concave-tariff/PROOF.md).

### Attainment

For each generator in the common order, partition the roles requesting
it into blocks of size `b` and one remaining block if needed. On each
block `B` use the coordinate projector `E_B` and the paid factor
`I + ([s]-[e])E_B`. In written product order, the entry on role `j`
is exactly the subsequence product `[w_j]`. All other entries have their
required diagonal or zero value. This is equality in the full group
algebra. The rightmost factor acts first. The construction uses just
the requested ranks, even when `b > r`.

### Mixed A3 example

Take adjacent swaps `s,t,u` in order and five targets
`(s, st, tu, su, stu)`. With cap two, their respective generator demands
are `(4,3,3)`. One attaining written word is

```text
s{1,2} s{4,5} t{2,3} t{5} u{3,4} u{5}
```

At `f(k)=k/(k+1)` this costs **11/3**. The pooled root-orbit bound is
`10/3`, because the ten inversions form one orbit and fit into five
parts of size two. Separate target-by-target calls cost `5`. The
separate cuts explain why the pooled lower bound is strict, and the
schedule saves `4/3` against separate calls.

## Evidence and reproduction

From the repository root with CPython 3.12.15, standard library only:

```sh
python3.12 -I -W error -B research/coxeter-common-order-tariff/verify.py
python3.12 -I -O -W error -B research/coxeter-common-order-tariff/verify.py
python3.12 -B scripts/research_package.py --package research/coxeter-common-order-tariff
python3.12 -B scripts/check_research_commit.py --package research/coxeter-common-order-tariff --check-source-history
```

Both direct commands print `PASS_COXETER_COMMON_ORDER_TARIFF`. They check
10,512 full formal A3 endpoints with a restored spectator, over all three
rank caps, all six generator orders and all ordered lists of up to three
non-spectator targets of that form. They also check 873 paid-factor cuts
using every two-by-two coupling over F2 and F3, including singular
matrices. The separate five-target example checks all six batch omissions
and reversal of the word. The written argument gives the universal result.

## Scope and credit

This contribution was produced by Nick Petalas. Benny Zhao's #6817 posed
the heterogeneous-target question and supplied the rank-capped model and
packing bound. ToikY's #6821 supplies the repeated A2 precedent with
selective cuts. Credit Adam Gall #1941 and Ross Richardson #1966/#1989 for
the separate-letter cut method, and David Arias #2738 and the retained
lineage for Coxeter tariffs. The new step is the coset cut for each simple
generator with simultaneous attaining batches for the common-order
heterogeneous family. [#6817](https://github.com/CrocSwap/integer-mult-bounds/pull/6817),
[#6821](https://github.com/CrocSwap/integer-mult-bounds/pull/6821),
[#2738](https://github.com/CrocSwap/integer-mult-bounds/pull/2738),
[advance claim](https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-19717633).

The theorem concerns the formal group-algebra tariff. Native
realization, physical scalar-map and movement costs, precision, compiler
closure, supplier admission and a new κ remain open. Independent human
and formal proof-kernel review are unclaimed.

## Open next questions

- For target orders that disagree on a noncommuting pair, what extra
price does chronology require? Start with `(st,ts)` and cap two.
The two individual generator cuts bound rank mass but do not encode
relative order.
- Can one physical rank-two primitive realize a formal batch at a net
saving? Start by specifying the boundary layout and pricing one
shared call including all routing and restoration. A supplier needs
that conversion before the formal price can enter its invoice.
36 changes: 36 additions & 0 deletions research/coxeter-common-order-tariff/SOURCES.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,36 @@
{
"schema_version": 1,
"base_commit": "e54523517d0f03e58352e647c403a5af3e525a57",
"files": [
{
"path": "certificates/selected-result.json",
"usage": "provenance",
"git_blob": "75d77b328ea02ded535e7f23aee5b1e55d0bb1a8",
"sha256": "154c3d30455d778fd3343b6b4851f0fa4665f0f7625b0adb3675e2bafbbf6bf1"
},
{
"path": "docs/research/agent-task-catalog.json",
"usage": "provenance",
"git_blob": "3404fbdd5e7529a0674eef26114b5071e4586cec",
"sha256": "ead6cdc9d0af6d58e8a49744d8ec8c7c15e13d2ba93c276c434bcf9b51b93a81"
},
{
"path": "research/a2-st-all-cap-tariff/PROOF.md",
"usage": "provenance",
"git_blob": "7f2279ec34893224361921529e7a7c9734405252",
"sha256": "76c25069149a699d9811c7e3dd92a9fd7271f271b5de87c9d215777b73b4999a"
},
{
"path": "research/coxeter-capped-concave-tariff/PROOF.md",
"usage": "provenance",
"git_blob": "00c633badd2f1848ec34b9893553161012e7a8f3",
"sha256": "1ef4cb87aee9ca82b9fe42a114b152e1241b7454d68dd49ea8d260fb91e28353"
},
{
"path": "research/coxeter-root-tariff/PROOF.md",
"usage": "provenance",
"git_blob": "c67d51701a3dccf1028f8213502979372dbd73b3",
"sha256": "2e20ac459ed1524fca510b58d2792004e44024fd96b9f775501c47786c8325f8"
}
]
}
79 changes: 79 additions & 0 deletions research/coxeter-common-order-tariff/TASK.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,79 @@
{
"schema_version": 1,
"task_id": "bit-new-register-model",
"priority_revision": 8,
"base_commit": "e54523517d0f03e58352e647c403a5af3e525a57",
"original_card": {
"id": "bit-new-register-model",
"question": "Can an explicitly larger or nonabelian register model escape the scoped three-register obstruction?",
"starting_points": [
"docs/research/research-family-status.md",
"research/bit-three-register-lower-bound/README.md",
"certificates/selected-result.json",
"research/bit-gl3-mixed-targets-length10/README.md",
"docs/research/community-efficient1000-review.md"
],
"methods": [
"Construct the smallest exact identity in a newly specified legal operation alphabet.",
"Derive a target-specific relaxation bound for that same explicit alphabet."
],
"acceptance_gate": "Name the changed hypothesis, all counted roles and a uniform identity; finite column tests alone do not certify it. A scoped local result may be retained without a complete supplier. Expanding it into further construction rounds requires the family-level budget and obligation gate in research-family-status.md; an optimistic passing relaxation alone is not acceptance.",
"first_step": {
"command": [
"python3",
"-B",
"scripts/research_selected_baseline.py"
],
"expected": "PASS_RESEARCH_BASELINE_BINDINGS_ONLY",
"inspect": "Resolve the current selected review, separate complex/ordinary BIT/admitted BIT/outer budgets, and compare the card against live handoffs before starting. Historical ceilings require exact source/model equivalence.",
"scope": "Source-binding orientation only; not a scientific replay or all-size proof. Materialize selected inputs if this checkout is sparse."
},
"first_experiment": "First compare the alphabet, role count and length with the fully replayed #1974 census and #2027 mixed-root tariff. Identify a changed premise and a legal native implementation; then price one smallest completed instance before searching larger words.",
"bottleneck_link": "A small-width completed saving changes the bit primitive; a finite-address identity alone is insufficient.",
"allowed_variations": [
"Choose a fixed larger register count or a specified nonabelian operation family."
],
"scope_change_triggers": [
"Remaining inside the already-excluded three-register abelian model.",
"Repeating the two-register GL3 census or a covered three-register length without changing a hypothesis; a formal group-algebra word without native conversion is a partial result."
],
"overlap_terms": [
"three-register",
"nonabelian",
"group algebra"
]
},
"scope_status": "changed",
"actual_scope": "Exact formal rank-capped concave tariffs for heterogeneous diagonal Coxeter targets that are subsequences of one order of distinct simple generators.",
"scope_change_reason": "Take the heterogeneous-target question left by Benny Zhao in PR #6817. This is a formal price question without a native implementation.",
"bottleneck_link": "A formal register-model side result. The matching rank lower bound identifies which shared batches could be worth implementing at a physical boundary.",
"budget_minutes": 60,
"stop_rule": "Stop at the first matched lower bound and complete formal endpoint. Leave conflicting orders and physical tariffs to later work.",
"outcome": "completed",
"claim_type": "Scoped written proof with exact finite corroboration.",
"evidence": "10,512 complete A3 formal endpoints and 873 arbitrary-coupling cut checks. The mixed five-target example costs 11/3, with a pooled lower bound of 10/3 and separate cost 5. Six omission controls and the reversed-order control reject.",
"author": "Nick Petalas",
"ai_assistance": "Independent human and formal proof-kernel review are unclaimed.",
"open_questions": [
{
"question": "What is the extra tariff for targets with conflicting noncommuting generator orders?",
"why_it_matters": "The common-order construction does not schedule opposite precedence constraints.",
"first_step": "Price the two-target tuple (st,ts) at cap two by using its full formal coefficients.",
"blocked_by": "The per-generator rank demands do not encode the relative order."
},
{
"question": "Can a physical rank-two primitive realize one shared batch at a net saving?",
"why_it_matters": "A saving in formal rank price enters the supplier only if it survives actual boundary costs.",
"first_step": "Specify an address layout and price one shared call, including routing and restoration.",
"blocked_by": "The native primitive and rank-to-recursive-cost conversion are unspecified."
}
],
"environment": "CPython 3.12.15 on macOS arm64; Python standard library.",
"freshness": {
"checked_utc": "2026-11-18T20:35:48Z",
"main": "e54523517d0f03e58352e647c403a5af3e525a57",
"priority_revision": 8,
"selected_source_pr": 7000,
"claim": "https://github.com/CrocSwap/integer-mult-bounds/discussions/1166#discussioncomment-19717633"
}
}
34 changes: 34 additions & 0 deletions research/coxeter-common-order-tariff/VERIFY.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,34 @@
{
"schema_version": 1,
"status": "ready",
"python": "3.12",
"requirements": [],
"inputs": [],
"commands": [
{
"argv": [
"{python}",
"-I",
"-W",
"error",
"-B",
"{package}/verify.py"
],
"timeout_seconds": 60,
"checks": "Full A3 formal endpoints, all two-role couplings over F2 and F3, the mixed cap-two tariff and seven rejecting controls."
},
{
"argv": [
"{python}",
"-I",
"-O",
"-W",
"error",
"-B",
"{package}/verify.py"
],
"timeout_seconds": 60,
"checks": "Full A3 formal endpoints, all two-role couplings over F2 and F3, the mixed cap-two tariff and seven rejecting controls."
}
]
}
Loading