MATHLIBANNEX / CANONICAL DECLARATION CARD

Root compression becomes scalar in norm

MathlibAnnex.CStarAlgebra.CAR.tendsto_norm_compressionError

theorem

Extends an exact rank-one compression identity on finite stages to a pointwise norm limit on the completed algebra.

Statement

Let be the completed CAR algebra with stage maps and root state . Put , where is the projection onto the first coordinate. Then, for every fixed , as .

Assumptions

The stages are linked by , and is their norm completion. The functional is the continuous root state. Each is a self-adjoint projection of norm at most one. The element is arbitrary but fixed while tends to infinity.

Conclusion

Writing , the theorem gives for every . The error is an element of ; its norm is the real-valued quantity that converges.

This is pointwise convergence in , not convergence in operator norm uniformly over the unit ball. It also does not assert norm convergence of the projections . The displayed projections have rank one at their defining finite stage; no rank-one assertion in an arbitrary representation of is used.

Proof route

If lies in a finite stage, rank-one compression gives whenever . For arbitrary , the projection norm bound and boundedness of give , uniformly in . Approximate the fixed by such a , and apply this estimate to .

Proof steps
  1. For , choose with , using density of the stages.

  2. For , the exact finite-stage compression identity gives .

  3. Linearity gives . Hence .

  4. This bound holds for every , which is the required norm limit.

Main citations

Lean source signature (exact)

theorem tendsto_norm_compressionError (x : Limit) :
    Filter.Tendsto (fun n => ‖compressionError n x‖) Filter.atTop (nhds 0)
In the source Mathematical meaning
x : Limit A fixed, arbitrary element of the completed CAR algebra.
compressionError n x The element , where and .
fun n => ‖compressionError n x‖ The real sequence using the C*-norm of .
Filter.atTop; nhds 0 Respectively in the natural-number order and convergence to the real number .
Filter.Tendsto (...) Filter.atTop (nhds 0) For this and each , there is such that implies . The threshold may depend on ; this is not uniform convergence over the unit ball.

Earlier published Card and PDF

Exact content identity

Declaration: MathlibAnnex.CStarAlgebra.CAR.tendsto_norm_compressionError

Accepted content SHA-256: c20769bed4b33c151af3f6e392b8c4571fb0e88f7a2955851bc1fe11c0e2c476

Accepted source guide SHA-256: 4f1a7868b114bf8dfae79ced8603c5ec487e2738937f61d31f25c49520c1ed82

Source commit: 437e6e46228bbb8e91211ebded349d7a30020e73

Back to top ↑