Keyboard shortcuts

Press ← or → to navigate between chapters

Press ? to show this help

Press Esc to hide this help

Finite-path rigidity and matrix decay

Abstract

Weighted finite-path rigidity, boundary observability, and spectral power decay.

Lemma 1.1 (Successive differences telescope).

Proof. Machine-checked in Lean as D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.sum_succ_sub_castSucc (✓ std3). ∎

Source. Repository-derived.

Commentary.

For every natural n and complex-valued function f on Fin (n + 1), the sum of successor differences is the last value minus the first value.

Definition 1.2 (Zero endpoint extension).

Formalization. D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.zeroExtend (✓ std3).

Source. Repository-derived.

Commentary.

For every m and complex vector z on Fin m, Fin.cases places zero at the left endpoint and Fin.snoc places zero at the right endpoint. The resulting vector has domain Fin(m + 2).

Lemma 1.3 (The interior values are unchanged).

Proof. Machine-checked in Lean as D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.extend_interior (✓ std3). ∎

Source. Repository-derived.

Commentary.

Interior coordinate i is coordinate i + 1 of the zero endpoint extension.

Lemma 1.4 (The right endpoint is zero).

Proof. Machine-checked in Lean as D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.extend_last (✓ std3). ∎

Source. Repository-derived.

Commentary.

The last coordinate of the extended vector is zero.

Theorem 1.5 (Weighted harmonic Dirichlet rigidity).

Proof. Machine-checked in Lean as D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.harmonic_dirichlet_zero (✓ std3). ∎

Source. Repository-derived.

Commentary.

For arbitrary m and positive velocities, equal adjacent weighted slopes force every interior complex coordinate to vanish. Finite induction makes all edge slopes equal. Their telescoping sum is zero; the strictly positive total velocity forces the common slope to zero.

Theorem 1.6 (A boundary zero propagates through the path).

Proof. Machine-checked in Lean as D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.boundary_zero_observability (✓ std3). ∎

Source. Repository-derived.

Commentary.

For m ≥ 1, a zero first interior coordinate and a recurrence propagating two adjacent zeros imply that the entire vector is zero. The left endpoint supplied by zeroExtend starts the induction.

Theorem 1.7 (Strict spectral bounds give real trajectory decay).

Proof. Machine-checked in Lean as D5/S3/Quantum/Dynamics/TridiagonalSweeps/FinitePathDynamics.real_mulVec_powers_tendsto_zero (✓ std3). ∎

Source. Repository-derived.

Commentary.

For m ≥ 1 and a real matrix N, every complex spectral modulus below one implies convergence of (N^k).mulVec u to zero for every real vector u. Gelfand’s formula gives an eventual geometric bound on matrix powers in the complex operator norm. Continuity of matrix-vector multiplication and of coordinatewise real parts yields the real trajectory limit.

References