Home·The theorem·The proof in one reading·Architecture·Units·Manuscript·Glossary·FAQ·Work items·Data·Method·Provenance·Changes
← I.1 Lattice, symmetries, gauge groups and the Wilson measurePart II·Detailed notesII.2 Marker algebra, polymer combinatorics and Gaussian estimates →

II.1. Field labels and their lattice representatives

Contents

1. What this unit proves 2

2. What it uses 4

3. Strategy 4

4. Steps 4

5. Where the difficulty lies 5

6. Notation 5

7. What remains unexplained 5

Code dependencies 5

Lean source 6

Additional source locations 6

Key statements 12

Part II — Data of the construction.

Describes local field labels and the lattice observables that represent them.

Status: edited. Condensed from the unit's detailed notes, which were checked against the linked Lean in two rounds and edited; the local limitations still apply. Status meanings.

Overview · All units · Detailed notes · ← Previous: I.1 · Next: II.2 →

Results · Inputs · Strategy · Proof steps · Notation · What remains unexplained · Code dependencies · Lean source · Key statements

Source inventory and conventions

Part: II, Data of the construction. Lean: work items G01–G07: YM/Labels/, YM/Source/, YM/Work/ (G01–G07), YM/Next/FOUNDATIONS/, YM/Push/RELAY_MARK/, YM/Extra/SOURCE_HISTORIES/, YM/Push/LOCAL_SPLITTING/, YM/Push/RELAY_DIFFERENCE/, YM/Planned/ATOM_GUARDED_LAYOUT/, YM/Further/LOCAL_LOG_JETS/, YM/LocalH/. Size: 10,805 lines in the source allocation.

1. What this unit proves

Throughout, \(G \subset U (n)\) is compact, \(n \geqslant 1\).

Proposition II.1.1 (labels and frames). Let \(\text{Tay}_j\) consist of \(4\)-tuples of symmetric \((j + 1)\)-linear maps \((\mathbb{R}^4)^{j + 1} \to \mathfrak{g}\) satisfying \(\sum_{\mu} z_{\mu} A_{\mu} (z, \ldots, z) = 0\). A jet is \(J = (J_j)_{j \geqslant 0}\); coordinates of \(J_j\) have weight \(j + 2\). For every \(d \geqslant 0\), the space \(V_d\) of \(\text{Ad} (G)\)-invariant complex jet polynomials of weight \(\leqslant d\) is finite-dimensional and determined by evaluation on real jets. It splits by weight and factor number \((w, f)\); conjugation and \(O (4)\) preserve grades. Its zero-factor part is \(\mathbb{C}1\). There exist nested real pure-grade bases \((b_i)\) of grades \((w_i, f_i)\) with \(b_0 = 1\). Status: definitions and theorems. Lean: frames.

Proposition II.1.2 (holonomies and sections). For every \(d, J, h, x, \mu\), the equation \(u' = u \hspace{0.17em} hA^{(d)}_{\mu} (h (x + te_{\mu}))\), \(u (0) = 1\), has a unique solution in \(G\), where \(A^{(d)}_{\mu} (z) = \sum_{j + 2 \leqslant d} J_{j, \mu} (z, \ldots, z)\). Its time-one endpoint \(U^{(d)}_{J, h}\) satisfies \(U_{J, h} = U_{\text{sc}_h J, 1}\) and \(U_{J, 0} = 1\). A section \((S, \Phi)\) maps coordinates linearly to link-entry polynomials on finite \(S\). Soundness requires signed-permutation-invariant \(S \ni 0\), gauge invariance, \(\Phi_{e_0} = 1\), \(\Phi_{\bar{c}} = \overline{\Phi_c}\), \(\Phi_c (s^{\ast} U) = \Phi_{s \cdot c} (U)\), and equality of the \(C^d\) jets of \(\Phi_c (U_{J, h})\) and \(p_c (\text{sc}_h J)\) at \(0\), for all \(c, J\). Status: definitions; ODE theorems. Lean: Hol.exists_ode; SectionData.Sound.

Proposition II.1.3 (primary sections). For every holonomy datum and coherent frames there are sound sections \(\Sigma_d\) whose column \(i\) is a polynomial in finitely many transported plaquette curvatures, every monomial having at least \(f_i\) factors. For every \(k, P \in \mathbb{N}\) there is \(A \geqslant 0\) such that, for every budget \(b\) of order \(k\), power \(P\), side and placement, column \(i\) is \(b\)-regular with norm \(\leqslant At_b^{f_i}\). For the fixed primary sections, nondecreasing \(\pi (D) \geqslant 1\) bounds these costs for \(d, k, P \leqslant D\), and \(c_d \geqslant 1\) bounds all column values globally. Status: theorem. Lean: Integrated.source_sections.

Proposition II.1.4 (readout). Fix holonomies, \(d\) and a frame. Every \(b\)-regular \(f\) on \(X \subset \Lambda_{\ell}\), with \(k_b \geqslant d\), has a unique read label \(p \in V_d\) at every \(o\): test jets agree through order \(d\) if \(10 \text{dist} (x, o) < \ell\) for all \(x \in X\); otherwise \(p = f (1) 1\). If also \(P_b \geqslant d\) and \(\lVert f \rVert_b \leqslant c\), then \(\lVert S_{t_b} \text{coord} (p) \rVert_1 \leqslant Cc\), with \(C \geqslant 1\) depending only on holonomies and frame. A sound placed section reads \(p_c\) if \(20 (\lvert x_{\mu} \rvert + 1) \leqslant \ell\) throughout its stencil. Status: theorem. Lean: exists_read; price_of_readout.

Proposition II.1.5 (equation sections). Let \(\mathcal{E}_d\) be the labels vanishing on all jets of radial potentials whose curvatures satisfy the free Bianchi and divergence equations, degree by degree. For every holonomy datum, \(d\) and frame, there exist a sound section with the slot structure and cost of II.1.3 and an automorphism \(R\) commuting with conjugation and signed permutations, such that \(\Phi_{\text{coord} (Rp)}\) is micro for every \(p \in \mathcal{E}_d\). Micro means continuous and, for some \(r, M\), orthogonal under every Wilson expectation with \(\ell \geqslant M\), every placement, and \(\beta \ne 0\), to every bounded measurable spectator using only links with both endpoints farther than \(r\) from the placement. The choices exist simultaneously along coherent frames. Status: unconditional theorem. Lean: Portfolio.build.

Proposition II.1.6 (observables). A record carries a history, holonomies, frames, sections, reference levels and transfer families. Define

\(\displaystyle O_{N, p} (v, o) = a_N^{- d} \Phi^{(d)}_{S_{1 / \theta_N} \tau^{- 1} v} (o) .\)

For all records this is linear, continuous, local and literally translation covariant; soundness gives gauge invariance. Under \(\text{Makes}\) and \(r (d) \leqslant N\), it respects reality and affine hypercubic pull-backs, satisfies \(O_{N, p} (a_{r (d)}^d e_0, o) = 1\), and has reference-level readout \(v\). Affine covariance is \(O (v, A_{s, b} o ; U) = O (s^{- 1} \cdot v, o ; A_{s, b}^{\ast} U)\). Here \(\text{Makes}\) requires selection, soundness, \(d + 3 \leqslant J (r (d))\), \(k_{b_m} \geqslant d + 1\), \(P_{b_m} \geqslant d\) for \(m \geqslant r (d)\), \(20 (\lvert x_{\mu} \rvert + 1) \leqslant \ell_{N, p}\) on \(S_d\) when \(r (d) \leqslant N\), column bounds \(\kappa_N \theta_N^{f_i}\) and based transfers. Status: theorems; \(\text{Makes}\) holds for VIII.1's family. Lean: Family.Makes; Family.O.

Proposition II.1.7 (transfers). With terminal linear automorphism \(\tau\), a based family for \((\Sigma, r)\) requires \(r \leqslant N\) and, for every \(v\) and \(r \leqslant m \leqslant N\), linear coordinates and remainders on fixed supports, regular remainders on \(X_i \ni 0\) with \(g_i (1) = 0\) and vanishing test derivatives through \(d\) when nonwrapping, and measurable gauge-invariant forced functions local in their supports and flags, vanishing unless eligible and carrying \(0\), and remainder bound \((\sqrt{w_m}, \kappa_m \lVert x_m^{\circ} \rVert_1)\). A bound \((\kappa, c)\) means summable regular and forced weighted masses \(\leqslant a\) and \(\leqslant \kappa b\), respectively, for \(a, b \geqslant 0\), \(a + b \leqslant c\), dominating regular norms and forced absolute values pointwise. It requires \(x_N = v\), zero terminal regular remainder, \(\mathcal{D}_N = \rho_N \Phi_{S_{1 / \theta_N} v}\), and \(\mathcal{D}_m = L_m^d T^{(m)}_{\beta_{N, p, m + 1}} \mathcal{D}_{m + 1}\). Also \(\lVert x_{m + 1} \rVert_1 \leqslant \kappa_{m + 1} \lVert x_m \rVert_1\), \(\lVert x_m \rVert_1 \leqslant \kappa_{m + 1} L_m^d \lVert x_{m + 1} \rVert_1\), \(x_r = \tau v\), \(\tau\) respects reality/symmetry, and \(x_m (a_N^d e_0) = a_m^d e_0\). For selected histories, \(\int \mathcal{D}_m (E_m (\tau^{- 1} v)) = a_m^d Z_{N, p} \langle O_{N, p} (v, 0) \rangle\). Status: specification, supplied by III.10; integral identity is a theorem. Lean: TransferData.Based.

Proposition II.1.8 (birth). Let \(\rho\) be lawful on \(\Lambda_{\ell}\) with \(w > 0\), threshold \(\theta\), \(o \in \Lambda_{\ell}\), \(Y = Y_o\), and \(\Sigma\) sound with \(\lvert \Phi_{e_i} (o ; \cdot) \rvert \leqslant B\), \(B \geqslant 1\). Suppose distinct reserved zero-activity forced addresses exist for every finite \(\varnothing \ne I \subseteq X\) and auxiliary index, with support \(X\), inventory \(I\) and everywhere-satisfied auxiliary test. If

\(\displaystyle B \theta^{- f_{\max}} \text{mass} (Y, 1) (e^{\lvert Y \rvert w} - 1) \leqslant \sqrt{w},\)

then, for every reservation index, there is linear \(E (v)\) with fixed supports, \(x = v\), \(g = 0\), Format and bound \((\sqrt{w}, \lVert v^{\circ} \rVert_1)\), satisfying \(\mathcal{D} (E (v)) = \rho \Phi_{S_{1 / \theta} v}\) pointwise. The forced part is precisely the large-field supplement. The guarded variant (Guard.summon, used at level N with w=w_N) additionally requires a selected history. It requires the auxiliary tests only when the support has exactly the prescribed flagged inventory, together with a separate zero-activity assumption. The section bound and fee condition are unchanged. Status: theorem. Lean: launched.

Proposition II.1.9 (numerical conditions). Let \(C_j \geqslant 1\), let \((\textsf{K}_j)\) be nondecreasing with \(\textsf{K}_j \geqslant 1\), and a stage profile satisfy \(j \leqslant k_j, P_j\), \(\pi (\max (k_j, P_j)) \leqslant \bar{\kappa}_j\), \((C_j + 1) \Omega_d \leqslant \textsf{K}_j\) for \(d \leqslant j\), and \(64 \textsf{K}_{j + 1}^2 \leqslant \bar{\kappa}_j\). There exist \(n_j, \beta_j^{\ast}\) such that every admissible level system of this profile with stage block factors \(\geqslant n_j\) and \(\hat{\beta}_m \geqslant \beta^{\ast}_{J (m + 1)}\) satisfies \(J (m) \leqslant k_{b_m}, P_{b_m}\), \(\pi (\max (k_{b_m}, P_{b_m})) \leqslant \kappa_m\), latch for \(d \leqslant J (m)\), and

\(\displaystyle 64 \textsf{K}_{J (m) + 1}^2 \leqslant \kappa_m, \qquad 10^5 \textsf{K}_{j + 1}^6 (1 + \kappa_{m + 1})^2 \leqslant L_m^{1 / 8},\)
\(\displaystyle \eta (1 + \kappa_{m + 1}) \Omega_d \leqslant L_m^{- 3 / 8}, \qquad (C_j + \eta \kappa_{m + 1}) \Omega_d \leqslant \textsf{K}_j,\)

where \(j = J (m + 1)\), \(\eta = L_m^{- 2 / 5}\), \(d \leqslant J (m)\). Latch requires sufficient side length and II.1.8's fee bound at every placement, with \(B = c_d\), \(\theta = \theta_m\), \(w = w_m\). Such plans exist by enlarging only stage constants whenever \(j \leqslant k_j, P_j\). For fixed \(d\) and support exponent, latch holds at sufficiently large coupling. These conditions and \(d + 3 \leqslant J (r (d))\) imply the section clauses of \(\text{Makes}\). Status: theorems. Lean: Doors.afford; Issue.refer.

Proposition II.1.10 (comparison). Diagonal sign changes, translations and conjugation preserve centred root tests of interior functions. If \(F_0\) is interior to a centred level-\(k\) cell with root \(\leqslant A\), and \(T\) has side divisible by that cell side and satisfies the seam inequality, every translate at distance at least three cell sides from its boundary has root \(\leqslant A\). Under \(\text{Makes}\), fix \(\gamma > 0\) and an extraction order. If finitely many eventually literal-translate families have centred roots \(\leqslant e^{- \gamma s_N}\) eventually at arbitrarily large fixed levels \(k \leqslant N\), with interior central support, then combinations with placement-independent coefficients eventually bounded by \(e^{\varepsilon s_N}\) for every \(\varepsilon > 0\) are close with bound \(y = e^{- \gamma s_N / 2} \to 0\): for every \(j, m \in \mathbb{N}\) there is \(k \geqslant m\) such that eventually \(k \leqslant N\) and, for every level-\(k\) link tiling \(T\) and \(o \in T. \text{margin} (0, \lfloor T. \text{side} / (8 (j + 1)) \rfloor)\), the combination is interior to cell \(0\) with root \(\leqslant y\). Status: theorems. Lean: offset; gather_close.

2. What it uses

Standard inputs: matrix ODEs, finite-group averaging, analytic inverse function theorem, polarization, norm equivalence, Haar integration by parts, reflection positivity.

3. Strategy

Invert a finite sampling of transported plaquette curvatures to realize jet labels. Readout reverses this construction. Link integration by parts realizes equation labels; marking reserved polymers restores the large-field complement of the insertion.

4. Steps

  1. Slots detect truncated jets: shrinking plaquette circulations recover curvature, hence radial potentials. Lean: Work.G03.answer.

  2. Taylor expansion of an equivariant inverse gives sections; each slot factor costs one threshold factor. Lean: Work.G05.answer; Work.G07.answer.

  3. Rescaled test links, polarization and finite-dimensional norm equivalence give uniform readout. Lean: coeff_bound.

  4. Force lifts avoid the varied link; a unitriangular change absorbs higher-grade errors. Lean: weaveBuild.

  5. Splicing the large-field supplement into reserved addresses gives exact birth. Lean: Attach.paid.

  6. Seam log-convexity extends centred bounds to offsets. Lean: offset.

5. Where the difficulty lies

Readout constants must avoid volume dependence. Equation lifts must retain independence of the varied link. Lawfulness alone supplies no reserved addresses. Higher grade orders factor number first, then weight. Stencils are finite but nonexplicit.

6. Notation

\(\text{sc}_h J = (h^{j + 2} J_j)_j\); \((S_t c)_i = t^{f_i} c_i\); \(p_c = \sum_i c_i b_i\); \(c^{\circ} = c - c_0 e_0\); \(Y = o + S\). \(a_m, \theta_m, w_m, \kappa_m, J (m)\) are history parameters; \(\Omega_d = 1 + 2 \sum_{s, i} \lVert s \cdot e_i \rVert_1\) sums over signed permutations and frame indices. \(\mathcal{D}\) is the first-order inserted density. Budget regularity means locality, measurability, gauge invariance and \(C^{k_b}\) exponential-coordinate regularity on the small-field set. The threshold satisfies \(t_b \in (0, 1]\). Its norm bounds all ordered left derivatives, including values, in directions \(\lvert A_e \rvert \leqslant t_b (1 + \text{dist} (e_-, o))^{P_b}\). Bounds include the small-field boundary. A slot is \(\text{Ad} (U_{\omega}) F_P\), where \(F_P\) projects \((U_P - U_P^{- 1}) / 2\) onto \(\mathfrak{g}\). Eventually means volume first, then mesh.

7. What remains unexplained

Identification of the auxiliary equation sections with limit null labels remains tentative. Compatibility with all later transfer estimates belongs to III.2 and III.10–III.15.

Detailed notes: the full account of this unit, with proof sketches for every step, is on the detailed notes page.

Code dependencies

Code dependencies (from the compiled Lean). These lists follow references between declarations allocated to units; they can include shared definitions, record fields and alternative constructions. They are not a mathematical reading order. Only unit pairs with at least 25 references are shown; see the counting rules.

References to: I.1, III.16, III.1, IV.1, III.10, III.2.

Referenced by: III.11, III.12, III.13, VII.2, VIII.1, III.2, VII.4, VII.6, III.10, III.6, IV.1, VII.1, VI.2, V.1, III.3, III.9, III.14, III.4, III.15, V.3, III.8, VII.3, VII.5, V.2.

Lean source

Source allocation: 10,805 attributed lines; 1,375 line-counted declarations; 1,893 assigned declaration names. Of the assigned names, 1,808 are in the compiled closure of the seventeen headline results. These quantities have different scopes.

The folders and result chains below are source inventories. Their line counts cover each whole group, including any share allocated to another unit; percentages use attributed lines plus one per assigned name. They need not sum to the unit total.

Additional source locations

Further pinned locations for the definitions and comparisons discussed in this unit.

Key statements

In the selected statement order. K- identifiers name this generated list; proposition and lemma numbers in the outline are separate. Each sentence is an automatically drafted summary of a Lean statement; follow its link for the exact hypotheses and conclusion. The outline status does not certify these summaries. Main result and key result describe importance, not correctness. Full names are in the statement table.

All units · Overview of the proof · Repository

← I.1 Lattice, symmetries, gauge groups and the Wilson measurePart II·Detailed notesII.2 Marker algebra, polymer combinatorics and Gaussian estimates →