Alphabeta Math
LemmaStatement: Literature-sourcedProof: AI-adaptedPipeline-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

Logical error and leakage add under code-preserving gate approximation

Statement

Let P be a code projection on a finite-dimensional Hilbert space V. Let W1,…,WG be ideal unitaries preserving ran⁡P, and let V1,…,VG be physical unitaries with ∥Vj−Wj∥op≤δj. If Vtot=VG⋯V1 and Wtot=WG⋯W1, then, with empty products equal to I and empty sums equal to 0, ∥Vtot−Wtot∥op≤∑j=1Gδj. Consequently the logical block error and leakage amplitude relative to the code action of Wtot are each at most ∑jδj.

Facts & Assumptions

Given: The code projection, the two finite lists of unitaries, and the individual bounds δj≥0.

[F1]

If W preserves the code, the logical block error and leakage amplitude are the compression and off-diagonal block norms. Logical block error and leakage amplitude

[F2]

A full-space operator approximation of size δ bounds both compressed logical error and leakage by δ. Block-target approximation bounds logical error and leakage

Proof

1.1givenF2algebra

If G=0, both products are the empty product I and the sum of errors is 0, so the full-space error is 0; [F2] then gives zero logical block error and zero leakage.

1.2givenalgebra

For G≥1, insert and subtract the mixed products to obtain the exact telescoping identity VG⋯V1−WG⋯W1=∑j=1GVG⋯Vj+1(Vj−Wj)Wj−1⋯W1. It follows by expanding the sum: consecutive terms cancel, leaving only VG⋯V1−WG⋯W1. Every outer product is unitary, so its operator norm is one and each summand has norm at most δj. The triangle inequality proves the displayed sum bound.

2.1F1F2step 1.1step 1.2algebra∎

The product Wtot preserves the code because each factor does. Apply [F2] with δ=∥Vtot−Wtot∥op; step 1.1 bounds this by ∑jδj using step 1.2 when G≥1 and step 1.1 when G=0. This gives the claimed logical error and leakage bounds, including on arbitrary ambient states before projection.

Depends on

Used by

Dependency tree · two levels

4 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources