Recognition: unknown
QKAN: quantum Kolmogorov-Arnold networks with applications in machine learning and multivariate state preparation
read the original abstract
We introduce quantum Kolmogorov-Arnold networks (QKAN), a quantum algorithmic framework inspired by the recently proposed Kolmogorov-Arnold Networks (KAN). QKAN inherits the compositional structure of KAN and is based on block-encodings, constructed recursively from a single layer using quantum singular value transformation. We demonstrate the algorithmic utility of QKAN in two applications. First, we introduce and analyze QKAN as a quantum learning model, treating the eigenvalues of block-encoded matrices as neurons and applying parametrized activation functions on the edges of the network. We show that QKAN is a wide-and-shallow neural architecture, where shallow depth is compensated by exponentially wide layers whenever efficient block-encodings of inputs are available. We further discuss how to parametrize and train QKAN using parametrized quantum circuits and quantum linear algebra subroutines. Second, we demonstrate that QKAN can serve as a multivariate quantum state-preparation protocol for functions with shallow compositional structure. We demonstrate this by efficiently preparing a multivariate Gaussian quantum state using a two-layer QKAN. Looking forward, we anticipate that QKAN's compositional and modular design will enable new applications in quantum machine learning and quantum state preparation.
This paper has not been read by Pith yet.
Forward citations
Cited by 6 Pith papers
-
Blind Catalytic Quantum Error Correction: Target-State Estimation and Fidelity Recovery Without A Priori Knowledge
Blind CQEC estimates target states from noisy outputs alone to recover quantum fidelity, achieving over 0.95 fidelity for small dimensions without noise-model knowledge via coherence maximization.
-
Blind Catalytic Quantum Error Correction: Target-State Estimation and Fidelity Recovery Without A Priori Knowledge
Blind CQEC estimates the target state from noisy data to recover fidelity without a priori knowledge, with a proven Lipschitz bound explaining linear correlation in fidelities and 3.4x error reduction in H2 VQE.
-
Merged amplitude encoding for Chebyshev quantum Kolmogorov--Arnold networks: trading qubits for circuit executions
Merged amplitude encoding reduces circuit executions in CCQKAN by a factor of n with 1-2 extra qubits and preserves trainability in numerical tests on networks and MNIST classification.
-
Catalytic Quantum Error Correction: Theory, Efficient Catalyst Preparation, and Numerical Benchmarks
Catalytic Quantum Error Correction recovers known target states from noisy copies with F > 0.96 using only eight copies by preserving coherent modes and applying a CPMG-Clifford-swap-test pipeline, bypassing magnitude...
-
The $\gamma_c$-Peak: Covariant Recovery on Four Organic Qubit Platforms
Four organic-material paths are outlined for magnetic-field-free quantum computing, with simulator benchmarks claiming significant error-correction gains and provable advantages over classical methods in toy-scale tasks.
-
The $\gamma_c$-Peak: Covariant Recovery on Four Organic Qubit Platforms
Simulations across four organic platforms show peak fidelity gains from covariant Petz recovery exactly at the entanglement-breaking threshold, with linear scaling in log d and significant advantages for several quant...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.