Pith. sign in
module module high

IndisputableMonolith.Foundation.PairKernelDiscreteGauss

show as:
view Lean formalization →

Develops the discrete double-entry structure for recognition currents: pair-flow antisymmetry alone forces global conservation and a discrete Gauss theorem (sum of divergences equals boundary flux). Cited by anyone in the pair-kernel or metric-edge gravity lane. Proofs are elementary finite-sum identities; gradient structure is not required.

claimA recognition current $F$ on pairs is antisymmetric when $F(i,j)=-F(j,i)$. Its discrete divergence is $\mathrm{div}\,F(i)=\sum_j F(i,j)$. Antisymmetry implies $\sum_i \mathrm{div}\,F(i)=0$ and the region identity $\sum_{i\in R}\mathrm{div}\,F(i)=$ boundary flux. Constant non-antisymmetric flows break conservation. Elementary postings are the basic antisymmetric ledger moves.

background

Recognition Science treats ledger exchange as a pair current on a discrete graph. The module sits under Door 2 / L0 locality (finite-range weight graph) from PairKernelLocality: the provenance lane that isolates on-site exclusion and the ratio bridge without yet committing to the full continuum bridge.

Double-entry is stated purely as antisymmetry: flow $i\to j$ equals minus flow $j\to i$. That is the bookkeeping content of "every debit has a matching credit" at the elementary current. A gradient $F(i,j)=\varphi_i-\varphi_j$ is one instance, but none of the conservation theorems need the gradient form.

Notation introduced here: IsAntisym for the antisymmetry predicate; divF for the discrete divergence; elementaryPosting for the basic antisymmetric move; counterexamples via constant flows that are not antisymmetric and therefore break global sum-to-zero.

proof idea

Definition-and-lemma module, not a single deep theorem. Antisymmetry is recorded as a predicate; divergence is the obvious site sum. Finite double-counting then gives: sum of an antisymmetric kernel over a Finset vanishes, hence total sum of divF is zero. The region form rewrites the interior sum as a boundary flux by the same pairing. Constant flows are exhibited as non-antisymmetric with nonzero total divergence, so conservation is sharp. Elementary postings are checked to be antisymmetric by direct evaluation. No analytic estimates; pure finite combinatorics.

why it matters in Recognition Science

Supplies the discrete conservation backbone that gravity analysis imports when reading strain currents on the Freudenthal patch. Downstream, MetricEdgeImage4D treats a finite linearized metric edge image: $F$ as strain current of a Mat4 perturbation on the sixteen-site patch, matching the frozen world metric-null plan. Without antisymmetry-level Gauss, edge-image conservation and boundary matching would be ad hoc.

In the foundation chain this is the double-entry half of the pair-kernel story (locality upstream, discrete Gauss here). It separates bookkeeping from dynamics: RCL and J-cost live elsewhere; here only the signed ledger identity is forced. That split keeps later continuum or continuum-limit claims from smuggling conservation in as a modeling choice.

scope and limits

used by (1)

From the project-wide theorem graph. These declarations reference this one in their body.

depends on (1)

Lean names referenced from this declaration's body.

declarations in this module (16)