Long Range Frequency Tuning for QML

TL;DR

This paper introduces a ternary grid initialization method for trainable-frequency quantum circuits, overcoming spectral gap limitations and enabling accurate frequency spectrum adaptation during training.

Contribution

The paper proposes a novel ternary grid initialization that ensures all target frequencies are near initialization points, improving training success in frequency tuning for quantum circuits.

Findings

Ternary initialization achieves median R^2 of 0.997 on synthetic benchmarks.

Standard initialization yields median R^2 of 0.18, showing poor frequency approximation.

Ternary initialization outperforms unary initialization and fixed baselines on real datasets.

Abstract

Angle-encoded variational quantum circuits admit a truncated Fourier series representation of their output, but approximating functions with maximum frequency $\omega_{\max}$ using fixed unary encoding requires $\mathcal{O}(\omega_{\max})$ encoding gates. Trainable-frequency (TF) circuits promise a reduction by learning the data-encoding prefactors alongside the ansatz parameters, adapting the accessible frequency spectrum to the target during training. We identify a practical barrier that prevents this promise from being realized: the prefactor gradient is suppressed by the spectral gap between the circuit's accessible frequencies and the target spectrum, independently of the ansatz parameters, confining gradient-driven prefactor movement to a narrow neighborhood of initialization. We propose \emph{ternary grid initialization} -- setting prefactors to $\{1, 3, 9, \ldots, 3^{k-1}\}$ -- which resolves this limitation by ensuring every target frequency within $[-\omega_{\max}, \omega_{\max}]$ lies within $\tfrac{1}{2}$ unit of a grid point at initialization, removing the spectral gap suppression by construction. On a synthetic benchmark with target frequencies shifted well beyond the standard initialization range, ternary initialization achieves median $R^2 = 0.997$ versus $0.18$ for unary initialization, with $100\%$ of runs achieving $R^2 > 0.95$ against $0\%$. CMA-ES with $20\times$ the evaluation budget reaches only $25\%$ success, confirming the limitation is a property of the optimization landscape rather than of gradient-based optimization specifically. Real-world validation on two benchmark datasets demonstrates consistent advantages over both fixed and trainable unary baselines.