How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Superstable regular continuum grand pressure and variational formula
Statement
Assume AC, the ultrafilter extension principle, DC, and HB. Let be superstable and regular as in Superstable regular pair interactions and configurational ensemble notation, and fix . Use the stationary finite-intensity law space , the specific relative entropy , the intensity , and the positional energy density defined in Superstable regular pair interactions and configurational ensemble notation and Local-Tame Topology. The existence and lower semicontinuity of under these hypotheses are established below. For free and periodic cubes define Then the limit exists, is independent of , and The minimizer set is nonempty and compact in the local-tame topology . This theorem concerns positional configurations only. If is instead normalized against Lebesgue measure with , then , so its logarithmic rate is .
Facts & Assumptions
Given: AC, the ultrafilter extension principle, DC, HB, a superstable regular pair potential, , and the unit-intensity Poisson reference.
The choice assumptions are carried explicitly because Compactness of Finite-Specific-Entropy Levels in the Local-Tame Topology uses AC, the ultrafilter extension principle, DC, and HB for compactness and sequential compactness.
Superstable regular pair interactions and configurational ensemble notation: superstable means a stable part plus a nonnegative repulsive core, regularity supplies the decreasing integrable lower and upper tail, and uses normalized unit-intensity Poisson measure.
Cell-quadratic lower bound from superstability: there are such that .
Compactness of Finite-Specific-Entropy Levels in the Local-Tame Topology: for every finite , the laws with are compact and sequentially compact in under the assumptions in [A1].
Spatial Mean-Information Theorem: if a stationary law is ergodic under all translations and has finite , then in for integer side lengths . The real-side extension used below is proved in step 1.3.
Local-Tame Topology: is convergence of expectations of every measurable local observable bounded by a constant times on some bounded set .
Free-cube grand-pressure limit for decreasing integrable tails: for the regular tail class here, the free-cube grand partition is finite and its free pressure limit exists.
Monotone convergence for the integral: nonnegative cell-count sums increase to their extended expectation.
Specific Relative Entropy Density gives the real-side cube entropy limit; Relative Entropy Variational, Chain, and Conditioning Rules and Relative Entropy of Probability Measures: finite-region entropy has the conditional chain rule, is nonnegative, and its variational formula applies to bounded measurable tests.
Georgii's Theorem 1 (1994, Section 3) states that for a superstable lower-regular pair potential, exists in for every stationary finite-intensity law and is lower semicontinuous in the local topology. Its hypotheses match [F1]–[F2]; the full 1994 proof was read. This supplies the energy-density existence interface, not a certification stamp.
Georgii's Theorem 2 (1994, Section 3, equations (2.20)–(2.22)) states that for a superstable regular potential and , the free and periodic normalized-Poisson partition functions have the same limiting rate , and the associated rate functional has compact local-tame sublevel sets. Its proof is in Sections 4–5; the complete source proof was read. The local Poisson/entropy inputs used there are supplied by [F2]–[F8].
Proof
Given: The assumptions and suppliers in Facts & Assumptions.
Georgii's Theorem 1 [L1] gives existence of and lower semicontinuity for every stationary finite-intensity law; by the Definition, this is . The source uses the same maximum norm in (2.3)–(2.7). Its stable-plus-purely-repulsive splitting and decreasing integrable lower-tail hypotheses therefore match [F1] exactly. Thus has exactly the domain and convention in the Statement.
The cell estimate [F2] and imply is lower semicontinuous and coercive on bounded- sets: the Poisson entropy variational inequality with bounded truncations of gives ; the pointwise square inequality gives . Hence bounded bounds , , and from above. A finite candidate exists: take independent Bernoulli occupancy of widely separated small cubes, at most one absolutely continuously distributed point per cube, and a uniform random lattice phase. Each tile has finite relative entropy to Poisson, and the summable regular tail bounds its absolute intertile energy per volume. Thus is proper. By [F3] and the choice assumptions [A1], its bounded sublevels are compact in .
To extend [F4] to real side lengths, fix its stationary ergodic with , put , and write and . Every such entropy is finite by Specific Relative Entropy Density. The chain rule [F8] factors -almost surely, where is the conditional density of the remaining region relative to its independent Poisson reference. Against the reference probability , is a density of mass one. Thus , and , so . The specific-entropy supplier gives and ; since , the right side divided by tends to zero. Together with [F4] this proves in .
For the finite-volume Gibbs tilt , direct integration gives Georgii's Theorem 2 [L2], applied with the present superstable regular interaction and , supplies the matching upper and lower empirical-field bounds and the free/periodic pressure limit. Its upper proof uses this entropy identity, block stationarization, and the superstable cell cutoff; its lower proof conditions finite blocks on bounded finite-energy events, inserts empty corridors, and applies the spatial likelihood limit. The inspected source hypotheses match [F1]–[F8]: the Poisson reference is independent unit intensity, is the local-tame topology [F5], entropy compactness is [F3], and the energy-density interface is [L1]. The theorem therefore yields for both boundary conditions, with compact variational sublevels.
The minimum is attained: a minimizing sequence lies in a finite sublevel by the sparse finite candidate in step 1.2; compactness [F3] gives a convergent subsequence and lower semicontinuity from step 1.1 gives a minimizer. The set of minimizers is the closed minimum level within that compact sublevel, hence nonempty and compact. This agrees with the compact-sublevel conclusion in [L2].
The unit-intensity Poisson law on has density relative to Lebesgue measure on the -point sector. Therefore the physical partition ; its log rate is . This is only a positional normalization conversion.
Depends on
The Axiom of Choice
The ultrafilter extension principle (UL/BPI)
The axiom of dependent choice: a relation in which every element is related to something admits an -indexed chain
The real dominated-extension principle as an additional hypothesis over ZF
Superstable regular pair interactions and configurational ensemble notation
Cell-quadratic lower bound from superstability
Free-cube grand-pressure limit for decreasing integrable tails
Poisson Configuration Reference
Poisson Reference Counts and Translation Invariance
Relative Entropy of Probability Measures
Relative Entropy Variational, Chain, and Conditioning Rules
Monotone convergence for the integral
Local-Tame Topology
Specific Relative Entropy Density
Compactness of Finite-Specific-Entropy Levels in the Local-Tame Topology
Spatial Mean-Information Theorem
Dependency tree · 2 levels
Sources
- H.-O. Georgii, Large deviations and the equivalence of ensembles for Gibbsian particle systems with superstable interaction
- H.-O. Georgii and H. Zessin, Large deviations and the maximum entropy principle for marked point random fields
- D. Ruelle, Superstable Interactions in Classical Statistical Mechanics
- D. Dereudre, Introduction to the theory of Gibbs point processes