Feedback-driven recurrent quantum neural network universality

TL;DR

This paper demonstrates that feedback-driven quantum recurrent neural networks are universal function approximators with efficient qubit scaling, enabling practical real-time quantum reservoir computing.

Contribution

It proves the universality of quantum recurrent neural networks with linear readouts and analyzes their approximation capabilities without the curse of dimensionality.

Findings

Quantum RNNs can approximate regular state-space systems efficiently.

Number of qubits grows logarithmically with inverse accuracy.

Quantum RNNs are universal with linear readouts.

Abstract

Quantum reservoir computing uses the dynamics of quantum systems to process temporal data, making it particularly well-suited for machine learning with noisy intermediate-scale quantum devices. Recent developments have introduced feedback-based quantum reservoir systems, which process temporal information with comparatively fewer components and enable real-time computation while preserving the input history. Motivated by their promising empirical performance, in this work, we study the approximation capabilities of feedback-based quantum reservoir computing. More specifically, we are concerned with recurrent quantum neural networks, which are quantum analogues of classical recurrent neural networks. Our results show that regular state-space systems can be approximated using quantum recurrent neural networks without the curse of dimensionality and with the number of qubits only growing logarithmically in the reciprocal of the prescribed approximation accuracy. Notably, our analysis demonstrates that quantum recurrent neural networks are universal with linear readouts, making them both powerful and experimentally accessible. These results pave the way for practical and theoretically grounded quantum reservoir computing with real-time processing capabilities.