Paper VII visual guide · Visual 36 of 40 · STRICT POSITIVITY AND THE GROUND GAP

A positive transfer determines its Hamiltonian

Traditional edition: Theorem II.10.2, Proposition II.10.5.

Start reading

Resources & source

Original plate in the monograph PDF · Original atlas source

Skip contents and start reading

On \({\mathcal H}={\mathbb C}^3\), take the strictly positive contraction

\[ T={\operatorname{diag}}(1,1/2,1/4),\qquad H=-\log T={\operatorname{diag}}(0,\log2,2\log2). \]

At unit Euclidean time, \(T=e^{-H}\). The self-adjoint logarithm is unique because the logarithm acts on positive spectral values. At integer time \(n\geq0\),

\[ T^n={\operatorname{diag}}(1,2^{-n},4^{-n}),\qquad P_0={\operatorname{diag}}(1,0,0),\qquad{\left\lVert T^n-P_0\right\rVert}=2^{-n}. \]

The gap \(\Delta=\log2\) determines the rate \(e^{-n\Delta}\) exactly in this example.

Three positive transfer eigenvalues have different decay rates.
Original drawing. Read the description below; expand the labels and data for selectable text.

Reading the drawing

The bars show the three diagonal transfer values at unit time. The annotation below gives their integer-time powers and the error from the ground projector. The unit eigenvalue persists while the other two decay.

Drawing labels and data 4 labels

Read the original labels as selectable text. Tables preserve grid relationships; other short labels are grouped compactly.

  • \(1\)
  • \(1/2\)
  • \(1/4\)
\[T^n:\quad1,\ 2^{-n},\ 4^{-n}\qquad\|T^n-P_0\|=2^{-n}\]

Open the full-size drawing

Strict positivity is an essential finite hypothesis

For \(T'={\operatorname{diag}}(1,0)\), the operator is positive semidefinite, but there is no finite self-adjoint \(H'\) on all of \({\mathbb C}^2\) with \(T'=e^{-H'}\): every finite eigenvalue of \(H'\) exponentiates to a strictly positive value. One may restrict to the positive support, but that changes the space on which the logarithm is defined.

For a family with nontrivial excited spectra, a uniform exponential rate needs a common positive lower bound on the gap. Ground multiplicity does not cause a problem: \(P_0\) selects the entire eigenspace at the minimum energy. For a Hamiltonian whose minimum is \(E_0\ne0\), the normalized transfer is \(e^{-t(H-E_0I)}\) before taking the limit.

Positivity is retained at several Euclidean times

For arbitrary vectors \(u_j\) and nonnegative integer times \(n_j\), the reflected kernel obeys

\[ \sum_{j,k}\langle u_j,T^{n_j+n_k}u_k\rangle =\left\|\sum_jT^{n_j}u_j\right\|^2. \]

This is a finite Gram factorization. The retained run checks the integer semigroup and diagonal ground error for \(n=0,\ldots,12\), and rejects a logarithm on the full space for the singular control. The written spectral argument justifies the logarithms and extends the functional calculus statement to all times.

Read this as

A positive definite transfer can be logarithmically reconstructed. Ground convergence needs a declared normalization and gap, while continuum reconstruction still needs its own structure.

Source and conversion record

The original atlas wording, equations, and drawing labels are retained. Drawings are views of the frozen monograph PDF. Text descriptions and table headings are added reading aids. This conversion does not rebuild TeX, rerun the examples, audit proofs or compile Lean.

Conversion checks and source hashes

Notation in Paper VII

Look up a symbol without leaving this chapter. Select text before opening help to search for it.

Search this work’s notation.

    Symbols can have different meanings in different passages. Read each result in its own context.

    Open the full notation index