Digital Quantum Reservoir Computing for ATM Time Series Prediction
2026-06-03 • Computational Engineering, Finance, and Science
Computational Engineering, Finance, and Science
AI summaryⓘ
The authors study a way to use small quantum computers to predict ATM cash demand over multiple future steps. They create a quantum model that encodes data into four quantum bits and only trains a simple classical part to make predictions. They test their method using simulations and a real quantum device, finding it does not beat traditional prediction methods on some error scores but does better on a measure that shows how well it captures patterns over time. This work helps understand how quantum computing might help with financial forecasts in the near future, while also showing where current limitations lie.
quantum reservoir computingATM cash demand forecastingtime series predictionparameterized quantum circuitsRidge regressionquantum measurementDynamic Time Warpingnear-term quantum devicesclassical vs quantum benchmarkingIQM Spark quantum processor
Authors
Chiara Vercellino, Giacomo Vitali, Valeria Zaffaroni, Francesca Cibrario, Emanuele Dri, Paolo Viviani, Olivier Terzo, Davide Corbelletto
Abstract
We investigate a digital quantum reservoir computing (QRC) framework for multi-step forecasting of automated teller machine (ATM) cash demand time series on near-term quantum devices. The proposed approach uses parametrized four-qubit reservoirs with a fixed structure exploiting partial measurement and reset, where temporal data is encoded in rotation angles. Training is restricted to a classical Ridge-regression readout. We systematically analyze the impact of the circuit ansatzë, reservoir memory, measurement-derived observables, and the execution backend on the forecasting performance. Experiments are performed with noiseless simulation, noise-aware emulation, and a real IQM Spark quantum processor. Although the QRC models do not outperform the classical Prophet benchmark in terms of Mean Absolute Error and Normalized Mean Squared Error metrics, they achieve more competitive results in Dynamic Time Warping metric, indicating a partial ability to capture temporal structure. These findings provide an empirical assessment of digital QRC for realistic financial forecasting and highlight both its current limitations and its potential on near-term quantum hardware.