def
definition
H_EnergyConservation
show as:
view math explainer →
open explainer
Read the cached plain-language explainer.
open lean source
IndisputableMonolith.Foundation.Hamiltonian on GitHub at line 92.
browse module
All declarations in this module, on Recognition.
explainer page
depends on
-
Energy -
IsTimeTranslationInvariant -
MetricTensor -
RRF -
TotalHamiltonian -
is -
is -
is -
MetricTensor -
is -
MetricTensor -
total -
total -
Hamiltonian -
MetricTensor
used by
formal source
89 TEST_PROTOCOL: Verify that dH/dt = 0 for stationary sections of the RRF potential.
90 FALSIFIER: Discovery of a time-translation invariant system where TotalHamiltonian
91 is not conserved. -/
92def H_EnergyConservation (psi : RRF) (g : MetricTensor) : Prop :=
93 IsTimeTranslationInvariant psi g →
94 (∀ t, TotalHamiltonian psi g t = TotalHamiltonian psi g 0)
95
96/-- **THEOREM: Energy Conservation (Noether)**
97 If the ledger constraints are time-translation invariant, the total recognition
98 energy (Hamiltonian) is conserved under RRF dynamics. -/
99theorem energy_conservation (h : H_EnergyConservation psi g) (psi : RRF) (g : MetricTensor) :
100 IsTimeTranslationInvariant psi g →
101 (∀ t, TotalHamiltonian psi g t = TotalHamiltonian psi g 0) := h
102
103/-- **HYPOTHESIS**: The Recognition Hamiltonian reduces to the standard physical
104 Hamiltonian in the small-deviation regime.
105 STATUS: EMPIRICAL_HYPO
106 TEST_PROTOCOL: Perform a Taylor expansion of H_rec around r=1 and compare to
107 standard field theory Hamiltonians.
108 FALSIFIER: Discovery of a deviation between H_rec and standard H that is large
109 enough to be measured in classical experiments. -/
110def H_HamiltonianEquivalence (psi : RRF) (g : MetricTensor) (x : Fin 4 → ℝ) : Prop :=
111 ∀ ε < 0.1, ∃ H_std : (Fin 4 → ℝ) → ℝ,
112 abs (HamiltonianDensity psi g x - H_std x) < ε^3
113
114/-- **THEOREM: Small-Deviation Hamiltonian Equivalence**
115 In the small-deviation regime (ε << 1), H_rec reduces to the standard
116 Hamiltonian of energy-based physics.
117 H ≈ ½(Π² + (∇Ψ)² + m²Ψ²). -/
118theorem hamiltonian_equivalence (h : H_HamiltonianEquivalence psi g x) (psi : RRF) (g : MetricTensor) (x : Fin 4 → ℝ) :
119 ∀ ε < 0.1, ∃ H_std : (Fin 4 → ℝ) → ℝ,
120 abs (HamiltonianDensity psi g x - H_std x) < ε^3 := h
121
122end Hamiltonian
papers checked against this theorem (showing 1 of 1)
-
Neural networks gain continuous depth by solving ODEs
"the adjoint sensitivity method (Pontryagin et al., 1962). This approach computes gradients by solving a second, augmented ODE backwards in time"