Alphabeta Math
DefinitionDefinition: AI-adaptedProof: Not applicablePipeline-generatedprecheck passverified 2026-10-22 (gpt-6.1-sol)
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.

Symmetric exclusion process under diffusive scaling

Definition

Assume the Axiom of Choice, meaning every family of nonempty sets has a choice function. Fix an integer d≥1 and an integer N≥3. Let ΛN=(Z/NZ)d, with each coordinate represented by 0,1,…,N−1, and let EN be the set of unordered nearest-neighbor bonds {x,x+ei}, where 1≤i≤d and addition is modulo N. The condition N≥3 makes this a simple graph with each bond listed once as x ranges over ΛN. The site graph is connected because any two coordinate tuples are joined by finitely many steps of +ei or −ei.

The configuration space is ΩN={η:ΛN→{0,1}}, ordered lexicographically by its binary occupation vector in the fixed site order; this labels it as a finite set {1,…,2Nd}. The occupation ηx is 1 for an occupied site and 0 for a vacant site. For b={x,y}∈EN, let ηb be the configuration obtained by swapping ηx and ηy.

The coordinates x/N have dimensionless spacing 1/N on the unit torus, and t is dimensionless diffusive time: it is the microscopic unit-rate bond clock accelerated by the factor N2. Thus the continuum heat equation below has normalized diffusion coefficient D=1.

Define a rate matrix LN on ΩN by giving each distinct state change η→ηb≠η rate N2 and setting the diagonal entry to the negative sum of the off-diagonal row. If the two occupations on a bond agree, swapping them produces no state change and contributes no off-diagonal transition. On a column observable f:ΩN→R, use the row-generator convention (LNf)(η)=∑ξ≠ηLN(η,ξ)(f(ξ)−f(η)); a row probability vector evolves by p′=pLN. The symmetric simple exclusion process (SSEP) is the uniformized finite-state chain with rate matrix LN and any specified initial probability vector. The finite-generator existence result supplies its all-pairs transition matrices etLN and càdlàg paths. Every jump conserves ∑x∈ΛNηx.

For k=0,1,…,Nd, the fixed-occupancy sector is ΩN,k={η∈ΩN:∑xηx=k}. When 0<k<Nd, this sector has at least two states and its transition graph is connected by the fixed-occupancy connectivity lemma. Setting every sector energy to 0, β=1, and each allowed transition rate to N2 makes this sector a finite reversible stochastic rate model with uniform stationary weights. The sectors k=0 and k=Nd are singleton constant chains. The full SSEP state graph is reducible across particle-number sectors; the generic finite rate-matrix construction, not the irreducible reversible-model definition, defines the full process.

Let Td=(R/Z)d with normalized Haar measure dx. For x∈ΛN, write x/N∈Td for the point with coordinates represented in [0,1). The empirical measure and its action on a smooth periodic test function G are

πtN:=N−d∑x∈ΛNηx(t) δx/N,⟨πtN,G⟩:=∫TdG dπtN=N−d∑x∈ΛNηx(t)G(x/N).

Here δy denotes unit point mass at y. Thus πtN is a finite positive measure of total mass at most 1, not necessarily a probability measure. For a smooth periodic G, define its discrete Laplacian on grid points by

ΔNG(x/N):=N2∑i=1d[G((x+ei)/N)+G((x−ei)/N)−2G(x/N)].

Depends on

Dependency tree · 2 levels

Sources