This study compares Quantum Reservoir Computing (QRC) with classical models such as Echo State Networks (ESNs) and Long Short-Term Memory networks (LSTMs), as well as hybrid quantum-classical architectures (QLSTM), for the nonlinear autoregressive moving average task (NARMA-10). We evaluate forecasting accuracy (NRMSE), computational cost, and evaluation time. Results show that QRC achieves competitive accuracy while offering potential sustainability advantages, particularly in resource-constrained settings, highlighting its promise for sustainable time-series AI applications.
Cite
@article{arxiv.2510.25183,
title = {Sustainable NARMA-10 Benchmarking for Quantum Reservoir Computing},
author = {Avyay Kodali and Priyanshi Singh and Pranay Pandey and Krishna Bhatia and Shalini Devendrababu and Srinjoy Ganguly},
journal= {arXiv preprint arXiv:2510.25183},
year = {2025}
}
Comments
6 pages, 1 table, 2 figures. Work conducted under QIntern 2025 (QWorld) with support from Fractal AI Research