Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

MK-1: an exactly simulated two-qubit ZZ-feature-map fidelity kernel, with machine-checked kernel properties

Benjamin Frohman

Published Nov 3, 2026• 0 citations
Publisher

Abstract

Technical note and code for MK-1, an exact statevector simulation of a two-qubit ZZ-feature-map fidelity kernel (compute-uncompute), following Havlíček et al. (2019). Documents a defect in the original script (its entangling gate commuted with CNOT, so the kernel factorised) and lists Lean 4 / Mathlib proofs (sorry-free, standard axioms only) that fidelity kernels from unitary feature maps satisfy K(x,x)=1, symmetry, 0≤K≤1, compute-uncompute = fidelity, and Gram positive semidefiniteness. No quantum-advantage claim.

DOI
10.5281/zenodo.23934621

Related papers

A Mathematica Package for Simulation of Quantum Computation

Vladimir P. Gerdt, Robert Kragler, Alexander N. Prokopenya

Lecture notes in computer science•Jan 1, 2009•9 citations

Statevector-to-Hardware Reconstruction of a Four-Qubit ZZ Quantum Kernel: A Single-Backend Case Study of Three Execution Jobs

Rostyslav Sipakov

Hardware noise and finite sampling perturb the fidelity estimates forming a quantum-kernel Gram matrix. We measured how far three hardware-reconstructed Gram matrices depart from an exact statevector reference for one frozen four-qubit ZZ feature map on N=24 indoor air-quality windows, executed on ibm_fez at 1024 shots per circuit in three single, non-interleaved jobs: baseline, dynamical decoupling alone, and gate twirling alone. All were complete, finite, and positive-semidefinite. Off-diagonal root-mean-squared error (RMSE) against the reference was 0.0878, 0.0864, and 0.0427; full-matrix centered kernel alignment (CKA) ranged 0.933-0.989 and the post hoc diagonal-excluded (U-centered) CKA 0.816-0.986. The gate-twirled job deviated least on every reported geometry axis; its baseline contrasts are deletion-stable for the Spearman, mean-absolute-error, RMSE, and full-matrix CKA diagnostics, while the Pearson and diagonal-excluded contrasts fall just below that convention. Dynamical decoupling was not separated from the baseline. The observed error exceeded both finite-shot reference scales, so, under those sampling-only models, sampling does not explain it. Centered kernel-target alignment did not track reconstruction fidelity and stayed at or below each label-permutation reference: implementation fidelity and task relevance are distinct diagnostic axes. All configuration-level statements describe three realized jobs on one backend; no mitigation-efficacy, classifier-superiority, forecasting, or quantum-advantage claim is made.

quant-pharXiv:2607.20377•Quantum Reports•Jul 23, 2026•1 citation

The matchgate even/odd split, kernel-checked: amplitudes are even-dimensional Pfaffians (2×2, 4×4, 6×6, and the recursive general object)

Rob Mitchell

Matchgates are classically simulable because their amplitudes are Pfaffians (signed sums over perfect matchings, equivalently square roots of determinants), and a determinant is a polynomial, computable in polynomial time. This note kernel-checks in Lean 4 + Mathlib the complete even/odd split of that correspondence over the complex numbers: the 2x2, 4x4 and 6x6 Pfaffians square to the determinant (the exact two-, four- and six-qubit amplitudes), the recursive Pfaffian defines the general even-2m object by Laplace expansion along the first row, and any odd-dimensional antisymmetric matrix has determinant 0 (det A = (-1)^n det A, proved structurally). The infrastructure det_fin_four / det_fin_five supplies the general 4x4 and 5x5 determinants Mathlib was missing. The general even-n identity det = Pf^2 (Cayley) and the full matchgate<->free-fermion correspondence remain the open rung, not claimed here. lake build green (9104 jobs), no sorry/axiom/admit.

•Nov 8, 2026•0 citations

New version of ZKCM, a C++ multiprecision matrix library usable for numerical studies of quantum information

Akira SaiToh

Computer Physics Communications•Feb 26, 2025•0 citations

Fractal dimension predicts quantum kernel collapse in angle-encoded data

Ana Paula Appel

Angle-encoded quantum kernels on tabular data collapse when the feature map is wider than the intrinsic dimension of the data. We propose the correlation fractal dimension D2 as an a priori qubit budget: encode D2 coordinates chosen by FD-ASE instead of the PCA-95% width or all E attributes. On nine data sets and a statevector simulator (n= 32), a one-layer ZZ fidelity kernel at q=D2 stays geometrically alive while the same kernel at the PCA-95% width has already collapsed. The budget is map-dependent: product-state and IQP maps overshoot it; a second ZZ layer undershoots it. Packed dense-angle and re-uploading encodings still live at the fractal q, but not when PCA-95% features are stacked onto those qubits. Shrinking the angle bandwidth moves the ZZ knee later; stretching it kills the kernel earlier. On IBM Quantum (ibm_fez, 256 shots, n=8) the one-layer ZZ kernel at the fractal width matches the exact kernel (MAE 0.021); past that width both hardware and simulator have collapsed. The ceiling is a property of the map-data pair at a stated bandwidth, not of the classical table alone.

quant-pharXiv:2609.00475•Sep 2, 2026•2 citations

The Kernel Quantum Probabilities (KQP) Library

Benjamin Piwowarski

In this document, we show how the different quantities necessary to compute kernel quantum probabilities can be computed. This document form the basis of the implementation of the Kernel Quantum Probability (KQP) open source project

arXiv:1203.6005•Mar 27, 2012•0 citations
Technical note and code for MK-1, an exact statevector simulation of a two-qubit ZZ-feature-map fidelity kernel (compute-uncompute), following Havlíček et al. (2019). Documents a defect in the original script (its entangling gate commuted with CNOT, so the kernel factorised) and lists Lean 4 / Mathlib proofs (sorry-free, standard axioms only) that fidelity kernels from unitary feature maps satisfy K(x,x)=1, symmetry, 0≤K≤1, compute-uncompute = fidelity, and Gram positive semidefiniteness. No quantum-advantage claim.
Hardware noise and finite sampling perturb the fidelity estimates forming a quantum-kernel Gram matrix. We measured how far three hardware-reconstructed Gram matrices depart from an exact statevector reference for one frozen four-qubit ZZ feature map on N=24 indoor air-quality windows, executed on ibm_fez at 1024 shots per circuit in three single, non-interleaved jobs: baseline, dynamical decoupling alone, and gate twirling alone. All were complete, finite, and positive-semidefinite. Off-diagonal root-mean-squared error (RMSE) against the reference was 0.0878, 0.0864, and 0.0427; full-matrix centered kernel alignment (CKA) ranged 0.933-0.989 and the post hoc diagonal-excluded (U-centered) CKA 0.816-0.986. The gate-twirled job deviated least on every reported geometry axis; its baseline contrasts are deletion-stable for the Spearman, mean-absolute-error, RMSE, and full-matrix CKA diagnostics, while the Pearson and diagonal-excluded contrasts fall just below that convention. Dynamical decoupling was not separated from the baseline. The observed error exceeded both finite-shot reference scales, so, under those sampling-only models, sampling does not explain it. Centered kernel-target alignment did not track reconstruction fidelity and stayed at or below each label-permutation reference: implementation fidelity and task relevance are distinct diagnostic axes. All configuration-level statements describe three realized jobs on one backend; no mitigation-efficacy, classifier-superiority, forecasting, or quantum-advantage claim is made.
Matchgates are classically simulable because their amplitudes are Pfaffians (signed sums over perfect matchings, equivalently square roots of determinants), and a determinant is a polynomial, computable in polynomial time. This note kernel-checks in Lean 4 + Mathlib the complete even/odd split of that correspondence over the complex numbers: the 2x2, 4x4 and 6x6 Pfaffians square to the determinant (the exact two-, four- and six-qubit amplitudes), the recursive Pfaffian defines the general even-2m object by Laplace expansion along the first row, and any odd-dimensional antisymmetric matrix has determinant 0 (det A = (-1)^n det A, proved structurally). The infrastructure det_fin_four / det_fin_five supplies the general 4x4 and 5x5 determinants Mathlib was missing. The general even-n identity det = Pf^2 (Cayley) and the full matchgate<->free-fermion correspondence remain the open rung, not claimed here. lake build green (9104 jobs), no sorry/axiom/admit.
Angle-encoded quantum kernels on tabular data collapse when the feature map is wider than the intrinsic dimension of the data. We propose the correlation fractal dimension D2 as an a priori qubit budget: encode D2 coordinates chosen by FD-ASE instead of the PCA-95% width or all E attributes. On nine data sets and a statevector simulator (n= 32), a one-layer ZZ fidelity kernel at q=D2 stays geometrically alive while the same kernel at the PCA-95% width has already collapsed. The budget is map-dependent: product-state and IQP maps overshoot it; a second ZZ layer undershoots it. Packed dense-angle and re-uploading encodings still live at the fractal q, but not when PCA-95% features are stacked onto those qubits. Shrinking the angle bandwidth moves the ZZ knee later; stretching it kills the kernel earlier. On IBM Quantum (ibm_fez, 256 shots, n=8) the one-layer ZZ kernel at the fractal width matches the exact kernel (MAE 0.021); past that width both hardware and simulator have collapsed. The ceiling is a property of the map-data pair at a stated bandwidth, not of the classical table alone.