SAQNN: Spectral Adaptive Quantum Neural Network as a Universal Approximator
This work addresses a foundational problem in quantum machine learning by providing a theoretical guarantee for QNN expressivity, which is incremental but important for the field.
The authors tackled the challenge of incomplete theoretical foundations for quantum neural networks (QNNs) by proposing a constructive QNN model that demonstrates the universal approximation property, achieving optimal parameter complexity for approximating Sobolev functions under the L2 norm.
Quantum machine learning (QML), as an interdisciplinary field bridging quantum computing and machine learning, has garnered significant attention in recent years. Currently, the field as a whole faces challenges due to incomplete theoretical foundations for the expressivity of quantum neural networks (QNNs). In this paper we propose a constructive QNN model and demonstrate that it possesses the universal approximation property (UAP), which means it can approximate any square-integrable function up to arbitrary accuracy. Furthermore, it supports switching function bases, thus adaptable to various scenarios in numerical approximation and machine learning. Our model has asymptotic advantages over the best classical feed-forward neural networks in terms of circuit size and achieves optimal parameter complexity when approximating Sobolev functions under $L_2$ norm.