Polynomial Signals
Exercise PLAN-516
primary source nl only. Current content-bound mathematical second pass accepted. No new Lean verification is claimed for this NL view.
Assumptions and conventions
- Single marked element; a = 1/sqrt(N) in the interval from zero to one
- T_{2k+1}(a) = (-1)^k sine((2k+1) arcsine(a)) for a in [0,1]
- Global phases do not change measurement probabilities
Question
View Grover's single-marked search through qubitization beyond the computational basis. Let the uniform state and the marked state define rank-one projectors, and let the reflection about the uniform state play the role of the accessing unitary while the two projector reflections play the roles of the input and output reflections. The encoded scalar is the overlap a = one over the square root of N = sine of the half-angle. Prove the identity that the odd-order flagged qubitization output, namely the output projector times the accessing unitary times (the full odd iterate) applied between the input and output projectors, equals the first-kind Chebyshev value of odd order 2k+1 at a, times the rank-one map from the uniform state to the marked state. Then use the fact that this odd Chebyshev value equals plus or minus the sine of (2k+1) times the arcsine of a to recover the Grover success amplitude, and identify the global phase relating this iterate to the standard Grover iterate.
Hint: mathematical route
Grover search is an odd Chebyshev singular value transformation of the scalar 1/sqrt(N): the flagged odd iterate is T_{2k+1}(a) times the uniform-to-marked map, equal to (-1)^k sine of (2k+1) half-angles, matching the Grover amplitude up to a global phase.
- The uniform-marked overlap gives the encoded scalar a = 1/sqrt(N) = sine of the half-angle.
- The flagged odd iterate equals T_{2k+1}(a) times (marked)(uniform adjoint).
- The special-argument identity turns T_{2k+1}(a) into (-1)^k sine of (2k+1) half-angles.
- The magnitude is the Grover success amplitude after k iterations.
- Letting this sine approach one recovers k of order sqrt(N).
- The qubitization iterate is minus the standard Grover iterate, giving a global phase (-1)^k.
Reference solution
The encoded operator is the one-by-one scalar a = sine of the half-angle = one over the square root of N, since the overlap of the uniform state with the marked state is a. The flagged odd iterate assembles, by the qubitization-beyond-the-basis construction, into the first-kind Chebyshev polynomial of odd order evaluated at the singular value a, acting as the rank-one operator sending the uniform state to the marked state: the output projector times the accessing unitary times the odd iterate between the two projectors equals T_{2k+1}(a) times (marked state)(uniform state adjoint). This is because the two-reflection iterate is a rotation whose (2k+1)-fold application produces, on the one-dimensional flagged overlap, exactly the odd Chebyshev amplitude, and the surrounding projectors extract the marked-from-uniform rank-one component. Now use the special-argument identity: for a in the interval from zero to one, the odd first-kind Chebyshev value T_{2k+1}(a) equals (minus one) to the k times the sine of (2k+1) times the arcsine of a. Since the arcsine of a is the half-angle, this is (minus one) to the k times the sine of (2k+1) half-angles. The magnitude of the flagged amplitude is therefore the absolute value of the sine of (2k+1) half-angles, which is exactly the Grover success amplitude for measuring the marked state after k iterations; letting the sine of (2k+1) half-angles approach one recovers the Grover query complexity (k of order the square root of N). The only difference from the standard Grover iterate is the sign (minus one) to the k: the qubitization iterate (reflection about uniform composed with the negated marked reflection) equals minus the standard Grover iterate, so its k-th power carries the global phase (minus one) to the k, which does not affect measurement probabilities. Hence Grover search is literally an odd Chebyshev singular value transformation of the scalar a, up to a measurement-irrelevant global phase.
Assessment rubric
- States the encoded scalar is a = 1/sqrt(N) = sine of the half-angle, as the uniform-marked overlap (1 weight)
- Establishes the flagged odd iterate equals T_{2k+1}(a) times the rank-one map from uniform to marked state (1 weight)
- Applies the identity T_{2k+1}(a) = (-1)^k sine of (2k+1) arcsine(a) to recover the Grover success amplitude (1 weight)
- Identifies the global phase (-1)^k relating this iterate to minus the standard Grover iterate, noting it is measurement-irrelevant (1 weight)
Graph-grounded prerequisites
These are support/source anchors, not an automatically extracted theorem-dependency proof.
- Block Encoding (frozen curriculum anchor; see graph registry)
- Harmonic analysis (frozen curriculum anchor; see graph registry)
- Approximation theory (frozen curriculum anchor; see graph registry)
Lean view and evidence boundary
No aligned Lean view has been admitted for this native core in this textbook checkpoint. This is not a claim that the underlying mathematics lacks a formal proof. Contributions must preserve this exact target contract and split.
Sources and provenance
- website/research/sources.json
lin-wiebe-2026-qasc
Source SHA-256:9f85f45f20ed5a46931ad0dec3b94ee09748bc56e39fae29b0aefe910440b0ae· MIT metadata only - Primary source
Contract: 8d28311cfe3fb9a671acd4693cf06c734bea72c058c9a50c349012ef138ca09f
Item digest: d16b9b79662aed1cc9a0d4dbaf86e932a72fa15e988bfa6581425efd1b448e4a
Graph snapshot: qcb-snapshot-ffa644f0da3031a1ce754b9a
Authored support steps: 6. This is not measured graph distance or model difficulty.