Pith. sign in
module module moderate

IndisputableMonolith.Geometry.DihedralCofactorFormula

show as:
view Lean formalization →

Coordinate-level formulas for the dihedral cosine of a tetrahedron edge from adjacent face normals. Opposite-vertex bookkeeping, edge vectors, cross-product normals, and the resulting numerator, squared denominator, and cosine are defined here. Downstream Regge-interior and dihedral-derivative modules import these identities. Mostly definitions plus elementary vector equalities linking dots and crosses to the geometric cosine.

claimFor a realized tetrahedron and edge $(i,j)$ with opposite vertices $(p,q)$, the two adjacent face normals are built as cross products of coordinate edge vectors from a base vertex. The geometric dihedral cosine is $\cos\theta = N/\sqrt{D}$ where $N$ is the inner product of the two normals and $D$ is the product of their squared norms (equivalently the normalized face-normal inner product).

background

Recognition Science geometry tracks tetrahedral dihedral angles both algebraically (Cayley-Menger cofactors) and metrically (realized edge vectors in Euclidean space). This module sits on the metric side. Upstream GramCayleyMenger isolates the bridge from the Euclidean Gram determinant of a realized tetrahedron to its Cayley-Menger determinant. Upstream DihedralCayleyMenger defines the tetrahedral dihedral cosine from Cayley-Menger cofactors for an edge $e=(i,j)$ with opposite vertices $(p,q)$.

Here the same edge is treated in coordinates. adjacentFaceOppositeVertices names the two remaining vertices opposite edge $(i,j)$. Coordinate edge vectors are differences of vertex positions; face normals are their cross products. Dot-product lemmas identify those vectors with the ambient inner product so that the geometric numerator, squared denominator, and cosine are ordinary Euclidean quantities.

The local setting is classical 3D vector geometry of a labeled tetrahedron, not yet the discrete curvature or Regge action. The module imports Mathlib cross and dot products and the two Cayley-Menger geometry modules above.

proof idea

Definition-heavy module, not a single theorem. It introduces opposite-vertex pairs for each edge, coordinate edge vectors (with equalities to base-point differences), and face normals as cross products. Lemmas then relate normal-normal and normal-self dots to the geometric numerator and squared denominator, and package the ratio as geometricDihedralCos. Cross-product forms of the numerator close the vector identities. No deep analytic argument: the content is bookkeeping plus standard $\mathbb{R}^3$ algebra so later modules can quote a named cosine.

why it matters in Recognition Science

Supplies the concrete face-normal cosine that the cofactor story must match and that analytic steps differentiate. Downstream AffineIndepInterior needs strict interior dihedral angles for the Regge closure program: its analytic core is that two linearly independent adjacent face normals have normalized dot product strictly in $(-1,1)$, and the remaining work is to prove those normals arise from the tetrahedron; this module is the source of those normals and their cosine. Downstream DihedralDerivatives isolates $d\theta=-(1/\sqrt{1-\cos^2\theta}),d(\cos\theta)$ once the Cayley-Menger cofactor cosine is differentiable; the geometric cosine defined here is the matching metric object.

In the broader RS geometry stack this is infrastructure between pure Cayley-Menger algebra and Regge-type curvature bookkeeping in $D=3$, not a forcing-chain landmark itself.

scope and limits

used by (2)

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

depends on (2)

Lean names referenced from this declaration's body.

declarations in this module (80)