QuantumComputinglib learn · inspect · formalize
Checked on this commit 4,524 public declarations commit 4e2c5b2c0be5 Build record

Structure before circuit tricks

Functor Hypergraph

Follow mathematical mechanisms across fields without confusing an analogy, a conditional construction and a formal implication.

All tails of a hyperedge are required together (AND). Incidence lines are not separate implications. Every transport here is curated or proposed; no Lean-certified categorical functor is asserted.

Dashed paths: curated/proposed incidence. AND junction: all listed hypotheses and tail mechanisms. Solid proof dependencies remain in the separate generated Lean graph.

Walsh phases + interference + nonzero branch

proposal Source-planning edge; independent conceptual review pending; no local transport theorem

\[-i(I-e^{-i\tau F_S})|+^n\rangle/2\approx\tau F_S|+^n\rangle/2\]
Mechanism
Combine a controlled diagonal phase supplier with ancilla interference, postselection and a normalized-error theorem.
Hypothesis map
Real function, source bit order, supplied coefficients, retained zero mode, explicit input and inverse access, small-angle/truncation bounds and nonzero accepted branch.
Conclusion map
An approximate amplitude state with a declared infidelity and success-dependent total cost, not an unconditional n-independent loader.
Failure boundary
Constant-term removal changes the controlled circuit; small filling fraction or poor sparsity can remove the advantage.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Efficient Quantum State Preparation with Walsh Series

primary-text-checked arXiv v3, 9 November 2023; checked 29 September 2026: Theorem 1, Corollary 1, Table I and Appendix B (B14–B19, B28).

Walsh Series Loader: epsilon is state infidelity. Smooth fixed-function single-attempt depth O(epsilon^-1/2), size O(n+epsilon^-1/2), success Theta(epsilon); expected depth O(epsilon^-3/2), expected size O((n+epsilon^-1/2)/epsilon). Constants retain function/derivative/nonzero-mass assumptions. Sparse variants require supplied coefficients and Hamming-weight bounds. Amplification also charges reflections and inverses.

Comment on arXiv:2307.08384: Efficient Quantum State Preparation with Walsh Series

primary-text-checked arXiv v1 full HTML, Sections II.2–II.3, Eq. (11), Fig. 2: controlled zero-order term; checked 29 September 2026.

Companion adversarial implementation review, not a refutation of the WSL theory. Retain the constant Walsh coefficient in a controlled phase circuit; use constant and constant-plus-parity regression cases.

Diagonal block + prepared input + success contract

proposal Source-planning edge; independent conceptual review pending; no local transport theorem

\[|\psi\rangle\longmapsto D|\psi\rangle/\|D\psi\|,\quad p=\|D\psi\|^2/\alpha^2\]
Mechanism
Specialize the existing BE-to-SP consumer to a diagonal filter; retain input preparation and successful-branch normalization as AND inputs.
Hypothesis map
Literal source dilation, bounded diagonal, known success sector, nonzero output, charged inverse/control and matched error metric.
Conclusion map
A conditional state and model-specific space–depth–accuracy/resource comparison.
Failure boundary
Neither a diagonal function description nor a compiled nearby fixed-size example certifies a scalable supplier; this is not SP-to-BE inclusion.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Efficient Quantum Circuits for Non-Unitary and Unitary Diagonal Operators with Space-Time-Accuracy trade-offs

primary-text-checked arXiv v4, 21 January 2025; checked 29 September 2026: II.1, III.1–III.2, IV.1 Eqs. (15)–(16), V.1 Eqs. (19)–(21), VI A, VIII.1 and Tables I–II. Figure 4 signal-sector and III.2/VIII.1 accounting ambiguities remain explicit source obligations.

Parallel scheduling trades ancillas for depth, not automatically total size. Table I uses spectral-norm error; Eq. (18) bounds normalized-block error delta, corresponding to total-scale times delta in the unscaled-operator convention. Exact-block success in source notation is ||D psi||^2/(alpha^2 dmax^2); alpha>1 controls the transformed-angle derivative in smooth approximation. No local source theorem or resolution of source ambiguities is claimed.

Quantum Algorithms for Scientific Computation

primary-text-checked 29 April 2026 public 447-page edition; PDF SHA256 56825000d53025dea4c2998dd6a061592b8fb2b9a446c3194c5f408183327599. First bounded source slice: Chapter 9 Section 9.1, printed pages 141–143, equations (9.2), (9.9), (9.10), (9.12); standing partial-application convention Definition 2.25, page 37. Parts I–IV remain curriculum anchors, not completed formalisation.

Peer Part IV curriculum anchor, reusing existing ASPBE state-preparation/block-encoding APIs. New Chapter 9 adapters, operator-norm approximation, scientific constructions and resource certificates still require separate production admission. No whole-section/book completion is claimed.

Exact function structure to bounded memory

curated-transport independent conceptual review pending; local Lean roots have their own build evidence

\[\{\text{degree and subdivision},\text{bit/branch contract}\}\Longrightarrow R\le2k+6\]
Mechanism
Keep local polynomial updates and branch routing together; neither alone proves the sampled source.
Hypothesis map
Endpoint jets, grid coordinate, boundary ownership and bit order agree.
Conclusion map
The finite core contraction equals the literal sampled Hermite function.
Failure boundary
A degree statement alone does not account for an arbitrary number of pieces.

Named Lean substrates below have their own exact signatures. They do not certify every sentence or proposed generalization on this page.

QuantumBlockEncoding.HermiteFiniteChain.sourceChain_contract

Locate the same declaration in the Lean graph

Normalize AND compile AND clean

curated-transport independent conceptual review pending; not a certified functor

\[\{\text{explicit TT},\ Z>0,\text{local compiler}\}\Longrightarrow U|0^{m+q}\rangle=|g_k\rangle|0^q\rangle\]
Mechanism
Compose the source, norm and primitive-circuit interfaces.
Hypothesis map
Real normalized scalar-boundary chain, rank bound, padded register layout and signed residual boundary.
Conclusion map
All target amplitudes and all non-clean sectors, plus actual primitive-list count.
Failure boundary
An existence-only TT representation or unknown normalizer is not a data-producing compiler.

Named Lean substrates below have their own exact signatures. They do not certify every sentence or proposed generalization on this page.

QuantumBlockEncoding.ConstructiveHermitePreparation.prepare_spec

Locate the same declaration in the Lean graph

SP + SELECT gives an LCU block

proposal independent review pending

\[P^\dagger\operatorname{SELECT}(U)P\rightsquigarrow A/\alpha\]
Mechanism
The AND node explicitly includes controlled operator access and unpreparation.
Hypothesis map
Square-root coefficient amplitudes, SELECT unitaries, adjoint access and phase convention.
Conclusion map
A specified projected matrix block, not merely one state.
Failure boundary
No arrow from an isolated prepared state to arbitrary A; state copies do not supply controlled U or U-dagger.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Circuit complexity of quantum access models for encoding classical data

primary-text-checked Results: circuit complexity lower bound; construction of LCU-based block-encoding; Methods: state preparation

Explicit access construction is not a free oracle. PREPARE together with SELECT and uncomputation can supply a block encoding; a single prepared state alone does not determine an arbitrary operator.

BE + input + overlap gives a state branch

proposal independent review pending

\[\Pi U_A(|0\rangle|\psi\rangle)=A|\psi\rangle/\alpha\]
Mechanism
Project the block, retain the accepted branch and account for success amplification.
Hypothesis map
Nonzero A psi, available input preparation and declared oracle inverses/controls.
Conclusion map
Normalized A psi conditional on success or an explicitly amplified approximation.
Failure boundary
Small overlap can dominate the cost; this is not an unconditional cheap reverse conversion.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Quantum State Preparation without Coherent Arithmetic

primary-metadata-checked Abstract; Physical Review Letters 136, 240603, published 18 June 2026

QET-based function preparation with few ancillas. Approximation, normalization and success probability remain separate costs.

Near-optimal ground state preparation

primary-metadata-checked Abstract: initial overlap, spectral-gap promise, energy information and lower bounds

Use the promised overlap/gap model; do not erase these costs in a generic strong-correlation claim.

Harmonic analysis to quantum evolution

proposal independent review pending

\[H_{\rm Sch}=D_p\otimes A_1-I\otimes A_2\]
Mechanism
The prepared auxiliary profile is one supplier; operator access and recovery remain separate.
Hypothesis map
Hermitian components, Fourier sign, finite grid and norm/recovery budget.
Conclusion map
Candidate Hamiltonian-access route for Schrödingerisation.
Failure boundary
SP certification alone proves neither the Hamiltonian block encoding nor end-to-end PDE accuracy.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

No external source is attached to this local mechanism note. It remains authored exposition, not a literature-priority claim.

Sampling envelope meets function structure

proposal independent review pending

\[\kappa_{\rm env}=C\|g\|_2/\|f\|_2\]
Mechanism
Search for an envelope with both a provable ratio bound and a constructive small representation.
Hypothesis map
Support domination, ratio degree/rank, phase access and charged reference preparation.
Conclusion map
A model-specific success and end-to-end cost target.
Failure boundary
A good classical envelope need not have a cheap coherent preparation; no universal cure for dimensionality.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Quantum Rejection Sampling

primary-metadata-checked Authors' publication page: quantum state generation, query characterization and matching lower bound

Prior art for coherent amplitude reweighting. Novelty must be in an explicit structured envelope, guarantees, implementation or model-matched bound, not in rejection sampling itself.

A coercivity mechanism, not an equality of samplers

proposal independent review pending

\[\text{invariance}+\text{quantitative dissipation}\Longrightarrow\text{a declared convergence rate}\]
Mechanism
Compare the role of a functional inequality in classical and quantum semigroups.
Hypothesis map
Different state spaces, noncommutativity, detailed-balance conventions and divergences remain explicit.
Conclusion map
A hypothesis-mapped conceptual mirror with candidate shared matrix/semigroup lemmas.
Failure boundary
Classical LSI or reversibility is not automatically a quantum KMS mixing certificate.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Samplinglib theorem-publication and conceptual-mirror protocols

primary-text-checked Author once; graph contribution; independent encoder–denoiser; unchanged historical audit debt

Design attribution. QuantumComputinglib has its own register, oracle, clean-ancilla and resource contracts; Samplinglib is not an imported Lean dependency.

Preparation structure plus measurement contract

proposal independent review pending

\[\Pr[|\widehat F-F|\le\epsilon]\ge1-\delta\]
Mechanism
Use the state description to propose a measurement witness; formal circuit proof and statistical evidence remain separate.
Hypothesis map
Experimental measurement access, noise, sample model, epsilon and delta.
Conclusion map
Candidate low-sample verification protocol for a specified family.
Failure boundary
An ideal Lean proof is not a hardware fidelity certificate.

No local transport theorem is bound to this record. Do not infer formal truth from its position in the atlas.

Hermite: dense tree to bounded-memory construction

add-nodeshortcutreorganisationbridge

Before

Generic amplitude tree retains one independent branch per sample; exponential gate count in the data-qubit count, not exponential ancillas.

After

Preserve the formula through Bernstein subdivision and bounded TT cores, then reuse a generic local-isometry compiler.

\[R=2k+6,\quad q=\lceil\log_2R\rceil,\quad G\le48mR^3\]

Preserved contract: Same discrete normalized Hermite state and clean-output convention; exact-real primitive gate model.

Curated repository contribution. Function-to-MPS and sequential preparation are prior art; the metadata does not certify global novelty, optimality or a finite-bit speedup.

Actual Git module/import delta: computed. This is not an elaborated proof-term delta.

Loading the computed Git delta; complete source data is linked below.

Solid lines: added imports. Dashed lines: removed imports. This is the changed module-import neighborhood, not a theorem proof graph or an automatically inferred novelty classification.

Added modules at the pinned contribution
  • module:QuantumBlockEncoding.AdjacentGivens
  • module:QuantumBlockEncoding.ConstructiveHermitePreparation
  • module:QuantumBlockEncoding.ConstructiveIsometryCompletion
  • module:QuantumBlockEncoding.ConstructiveIsometryLocal
  • module:QuantumBlockEncoding.ConstructiveTensorTrain
  • module:QuantumBlockEncoding.ConstructiveTensorTrainCompiler
  • module:QuantumBlockEncoding.ConstructiveThinLQ
  • module:QuantumBlockEncoding.GrayBasis
  • module:QuantumBlockEncoding.GrayGivensCompiler
  • module:QuantumBlockEncoding.HermiteBernstein
  • module:QuantumBlockEncoding.HermiteBoundaryInjection
  • module:QuantumBlockEncoding.HermiteCutRank
  • module:QuantumBlockEncoding.HermiteFiniteChain
  • module:QuantumBlockEncoding.HermiteFiniteNorm
  • module:QuantumBlockEncoding.HermiteIntervalMass
  • module:QuantumBlockEncoding.HermitePolynomialPreparation
  • module:QuantumBlockEncoding.HermitePolynomialResources
  • module:QuantumBlockEncoding.HermiteSampleStructure
  • module:QuantumBlockEncoding.HermiteTransferCores
  • module:QuantumBlockEncoding.MatrixProductChain
  • module:QuantumBlockEncoding.PrimitiveDepthBound
  • module:QuantumBlockEncoding.PrimitiveWireRename
  • module:QuantumBlockEncoding.RealIsometryCompletion
  • module:QuantumBlockEncoding.RectangularGivens
  • module:QuantumBlockEncoding.SelectedRyPlane
  • module:QuantumBlockEncoding.SelectedRyTrace
  • module:QuantumBlockEncoding.SequentialBondPreparation
  • module:QuantumBlockEncoding.SequentialPrimitiveAssembly
  • module:QuantumBlockEncoding.StoredBernstein
  • module:QuantumBlockEncoding.StoredGivens
  • module:QuantumBlockEncoding.StoredIsometryCompletion
  • module:QuantumBlockEncoding.StoredRectangularGivens
  • module:QuantumBlockEncoding.StoredTensorTrain
  • module:QuantumBlockEncoding.StoredThinLQ
  • module:QuantumBlockEncoding.TensorTrainCanonical
  • module:QuantumBlockEncoding.TensorTrainLocalCompiler
  • module:QuantumBlockEncoding.TensorTrainNormEnvironment
  • module:QuantumBlockEncoding.TensorTrainPrimitivePreparation
  • module:QuantumBlockEncoding.TensorTrainSchedule
  • module:QuantumBlockEncoding.TensorTrainWord
  • module:QuantumBlockEncoding.ThinLQ

Download baseline/head and exact module-import delta