Quantum Recurrent Neural Networks for Sequential Learning

By Y. Li et al

Published on Feb. 7, 2023

Read the original document by opening this link in a new tab.

Table of Contents

1. Introduction
2. Quantum recurrent block
3. Data encoding
4. Ansatz
5. Performance analysis
6. Conclusion

Summary

This paper introduces a novel Quantum Recurrent Neural Network (QRNN) model for sequential learning. The QRNN is designed to overcome the limitations of classical recurrent neural networks when applied to quantum data. The proposed QRNN architecture utilizes quantum recurrent blocks (QRBs) in a staggered manner to enhance efficiency on near-term quantum devices. Experimental results demonstrate superior performance in predicting sequential data such as meteorological indicators, stock prices, and text categorization. The study provides insights into the potential applications of QRNNs in the near future.