Quantum recurrent neural networks perform similarly to classical models on time series

Benchmarking memory and nonlinear processing capacity of quantum recurrent neural networks for time series prediction

Machine Learning

Summary

Predicting future values in a sequence, like weather or stock prices, requires remembering past information and handling complex patterns. The authors tested special quantum-based neural networks called QRNNs against classical ones to see how well they remember information and learn nonlinear patterns. They found quantum models can perform similarly but classical ones handle larger memory sizes more reliably. Quantum models show promise if starting parameters are tuned well, but training them gets harder as they grow bigger.

What this means in practice

Authors

José Daniel Viqueira, Andrés Gómez, David Mera

Abstract

Quantum machine learning has emerged as a promising approach to time series prediction. While quantum reservoir computing has been widely explored for this setting, variational quantum models tailored to sequential data remain comparatively understudied, largely because they are difficult to train. In this work, we present a rigorous benchmark of variational quantum recurrent neural networks (QRNNs) on time series prediction tasks, focusing on two key capabilities: memory and nonlinear processing. We introduce a task-driven methodology that quantifies model capacity, enabling an intuitive and fair comparison across architectures, including classical recurrent baselines. We evaluate the proposed framework on two generic benchmark tasks and on a synthetic dataset describing a chaotic system. Overall, the quantum models achieve performance comparable to their classical counterparts, but classical models scale better with the memory size within the range studied. Notably, one QRNN variant shows higher average capacity on the task designed to jointly probe memory and nonlinear capacity. Moreover, our results show that quantum models offer better capacity when we are able to find a good parameter initialization. However, the increasing sensitivity to initialization leads to a lower probability of finding good solutions as the model size increases. These results provide a principled starting point for designing richer quantum recurrent architectures and for assessing where they may offer practical benefits.