pith. machine review for the scientific record. sign in
theorem proved tactic proof high

second_law_one_statement

show as:
view Lean formalization →

The declaration proves that the excess recognition free energy H(n) = F_R(R^n q0) - F_R(gibbs equilibrium) is nonnegative for every n and strictly antitone along any J-descent trajectory. Researchers deriving the second law from recognition primitives would cite this as the Lyapunov form of the RS second law. The proof is a direct two-part split that invokes the variational inequality free_energy_ge_equilibrium for the lower bound and the monotonicity lemma free_energy_le_of_le for the antitone property, both discharged by linarith.

claimLet sys be a recognition system, X : Ω → ℝ a cost function, R a J-descent operator fixing the Gibbs equilibrium π = gibbs_measure(sys, X), and q₀ an initial probability distribution. Define H : ℕ → ℝ by H(n) := recognition_free_energy(sys, (R.step)^n q₀, X) − recognition_free_energy(sys, π, X). Then H(n) ≥ 0 for all n and H is antitone.

background

A RecognitionSystem fixes a positive recognition temperature T_R. The Gibbs measure π = gibbs_measure(sys, X) is the unique equilibrium distribution that minimizes recognition free energy for the given cost function X. A JDescentOperator on π is a structure whose step map fixes π and satisfies kl_non_increasing: D_KL(step q ‖ π) ≤ D_KL(q ‖ π) for every distribution q (see the structure fields fixes_equilibrium and kl_non_increasing).

proof idea

The tactic proof introduces the local definition of H and refines the conjunction. The left conjunct (nonnegativity) is obtained by applying free_energy_ge_equilibrium to the evolved distribution at step n and discharging the resulting inequality with linarith. The right conjunct (antitone) is obtained by applying free_energy_le_of_le to the free-energy values themselves and again using linarith on the differences.

why it matters in Recognition Science

This supplies the single-statement Lyapunov form of the second law and is the direct parent of secondLawCert_inhabited, which inhabits the SecondLawCert certificate. It completes the free-energy monotonicity derivation inside the module that assembles the second law from the J-descent property of the recognition operator. In the broader framework it realizes the monotonicity step that follows from T5 (J-uniqueness) and T7 (eight-tick octave) along the forcing chain.

scope and limits

formal statement (Lean)

 300theorem second_law_one_statement (sys : RecognitionSystem) (X : Ω → ℝ)
 301    (R : JDescentOperator (gibbsPD sys X)) (q₀ : ProbabilityDistribution Ω) :
 302    let H : ℕ → ℝ := fun n =>

proof body

Tactic-mode proof.

 303      recognition_free_energy sys (R.evolve n q₀).p X -
 304        recognition_free_energy sys (gibbs_measure sys X) X
 305    (∀ n : ℕ, 0 ≤ H n) ∧ Antitone H := by
 306  intro H
 307  refine ⟨?_, ?_⟩
 308  · intro n
 309    show 0 ≤ recognition_free_energy sys (R.evolve n q₀).p X -
 310        recognition_free_energy sys (gibbs_measure sys X) X
 311    have := free_energy_ge_equilibrium sys X R q₀ n
 312    linarith
 313  · intro n m hnm
 314    show recognition_free_energy sys (R.evolve m q₀).p X -
 315        recognition_free_energy sys (gibbs_measure sys X) X ≤
 316        recognition_free_energy sys (R.evolve n q₀).p X -
 317        recognition_free_energy sys (gibbs_measure sys X) X
 318    have hF :
 319        recognition_free_energy sys (R.evolve m q₀).p X ≤
 320          recognition_free_energy sys (R.evolve n q₀).p X :=
 321      free_energy_le_of_le sys X R q₀ hnm
 322    linarith
 323
 324/-! ## §7. The Clausius / entropy form -/
 325
 326/-- **THE SECOND LAW — CLAUSIUS / ENTROPY FORM.**
 327
 328    If the expected cost `⟨X⟩` is conserved along the J-descent evolution
 329    (i.e. the system is energetically isolated), then the recognition entropy
 330    `S_R(qₙ) = −Σ qₙ(ω) log qₙ(ω)` is monotone non-decreasing in `n`.
 331
 332    This is the classical statement `ΔS ≥ 0` for an isolated system,
 333    derived from the free-energy form `F = ⟨X⟩ − T_R · S_R`. -/

used by (1)

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

depends on (26)

Lean names referenced from this declaration's body.