Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2603.10707

Hybrid Photonic Quantum Reservoir Computing for High-Dimensional Financial Surface Prediction

Fyodor Amanov, Azamkhon Azamov

quant-phSubmitted Mar 11, 2026• Announced Mar 12, 2026• 1 citation
arXivHTMLPDF

Abstract

We propose a hybrid photonic quantum reservoir computing (QRC) framework for swaption surface prediction. The pipeline compresses 224-dimensional surfaces to a 20-dimensional latent space via a sparse denoising autoencoder, extracts 1,215 Fock-basis features from an ensemble of three fixed photonic reservoirs, concatenates them with a 120-dimensional classical context, and maps the resulting 1,335-dimensional feature vector to predictions with Ridge regression. We benchmark against 10 classical and quantum baselines on six held-out trading days. Our approach achieves the lowest surface RMSE of~$0.0425$ while maintaining sub-millisecond inference. The quantum layer has zero trainable parameters, sidestepping barren plateaus entirely. Variational quantum methods (VQC, Quantum LSTM) yield negative $R^{2}$ on test data, confirming that fixed quantum feature extractors paired with regularised readouts are more viable for low-data financial applications.

Comments
19 pages, 10 figures

Related papers

Photonic quantum reservoir computing on real financial surfaces: identifying the structural ceiling of near-random-walk data

Oishik Kar, Aditya Raj, Aswath Babu H

Abstract Quantum machine learning is increasingly proposed for financial time series, yet practitioners have no way to predict whether quantum reservoir computing (QRC) will outperform classical baselines on a given dataset before committing resources. We propose a pre-experimental screening framework, built on persistence R 2 and principal component analysis (PCA) variance decomposition, that makes this prediction possible. We validate the framework using a 10-mode Boson sampling circuit (252 features) on three market datasets (US Treasury yields, CBOE VIX, crude oil) and a diagnostic nonlinear control. On the market data, the screening tool correctly predicts a structural ceiling : because financial surfaces follow near-random-walk dynamics (persistence R 2 > 0.95 , 3 PCA components), QRC collapses to the performance of classical Ridge regression ( R 2 = 0.9797 vs 0.9798 on Treasuries, with persistence at 0.9793 ). On the nonlinear control, the framework correctly predicts that a nonlinear feature map will outperform linear Ridge. Running the actual Boson sampling QRC on this control confirms the prediction: the quantum circuit achieves R 2 = 0.816 ± 0.007 , cleanly outperforming classical Ridge ( R 2 = 0.77 ) on a dataset where persistence completely fails ( R 2 = − 1.64 ). The proposed screening test can be computed from the raw data alone, before any quantum circuit is built.

Physica Scripta•Oct 15, 2026•0 citations

Learning Temporal Patterns in Financial Time Series: A Comparative Study of Quantum LSTM and Quantum Reservoir Computing

Danyal Maheshwari, Gerhard Hellstern, Martin Zaefferer +2 more

This study explores quantum and classical hybrid architectures for financial time-series fore casting, focusing on Quantum Long Short-Term Memory (QLSTM) networks and Quantum Reservoir Computing (QRC), using univariate and multivariate lag structures on real financial data. We assess how lag embeddings affect predictive accuracy and robustness. Data are en coded into quantum states via amplitude encoding, enabling efficient representation of normalized lagged observations under realistic qubit constraints. The recurrent dynamics of QLSTM and the reservoir of QRC are implemented as parameterized quantum circuits, while classical optimizers train the readout and, where applicable, variational circuit parameters. We benchmark quantum models against classical LSTM and reservoir computing using common error like metrics. Our results show that, with suitable lag selection and amplitude encoding, quantum-enhanced archi tectures match classical baselines in univariate settings and can modestly outperform them in multivariate regimes with correlated inputs, where expressive encodings are most beneficial.

quant-pharXiv:2605.02656•May 5, 2026•1 citation

Quantum Recurrent Architectures for Financial Time Series Prediction

Danyal Maheshwari, Gerhard Hellstern, Martin Zaefferer +2 more

This study evaluates quantum classical hybrid recurrent architectures for financial time-series forecasting: a Quantum Long Short-Term Memory (QLSTM) network and Quantum Reservoir Computing (QRC) with lag-based inputs. Classical lag vectors are encoded into quantum states via amplitude encoding, which compactly represents normalized observations but requires nontrivial state preparation; the QLSTM gates and the QRC reservoir are implemented with parameterized quantum circuits, while classical optimizers train the interface and readout layers. On a monthly revenue series extended to 1056 months by a Gaussian-process--hidden-Markov generator, with a fixed chronological 70/15/15 split and one-step-ahead forecasts, we compare both models against a full-size LSTM, a classical control matched in parameter budget and interface architecture, a small plain LSTM, and classical reservoir computing, over ten seeds and against persistence, seasonal-naive and SARIMA benchmark forecasts. The QLSTM, with 2,105 parameters, is statistically indistinguishable from an LSTM 126 times larger, but equally indistinguishable from the matched classical control and from a plain LSTM of the same budget, so its parameter efficiency cannot be attributed to the quantum circuits. The quantum reservoir is worse than a classical reservoir of equal feature dimension in all ten seeds and worse than SARIMA, whereas all recurrent models and both classical reservoirs improve on SARIMA by about 2\% in RMSE. In a 24-month rolling-origin backtest the classical LSTMs degrade more slowly than the QLSTM. We find no evidence of a quantum accuracy or parameter-efficiency advantage in this low-dimensional setting.

arXiv:2611.15298•Nov 11, 2026•0 citations

Digital Quantum Reservoir Computing for ATM Time Series Prediction

Chiara Vercellino, Giacomo Vitali, Valeria Zaffaroni +5 more

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.

quant-pharXiv:2606.04686•Jun 4, 2026•0 citations

Image Denoising via Quantum Reservoir Computing

Soumyadip Das, Luke Antoncich, Jingbo B. Wang

Quantum Reservoir Computing (QRC) leverages the natural dynamics of quantum systems for information processing, without requiring a fault-tolerant quantum computer. In this work, we apply QRC within a hybrid quantum classical framework for image denoising. The quantum reservoir is implemented using a Rydberg atom array, while a classical neural network serves as the readout layer. To prepare the input, images are first compressed using Principal Component Analysis (PCA), reducing their dimensionality to match the size of the atom array. Each feature vector is encoded into local detuning parameters of a time-dependent Hamiltonian governing the Rydberg system. As the system evolves, it generates nonlinear embeddings through the measurement of observables across multiple time steps. These temporal embeddings capture complex correlations, which are fed into a classical neural network to reconstruct the denoised images. To evaluate performance, we compare this QRC-assisted model against a baseline architecture consisting of PCA followed by a dense neural network, trained under identical conditions. Our results show that the QRC-based approach achieves improved image sharpness and similar structural recovery compared to the PCA-based model. We demonstrate the practical viability of this framework through experiments on QuEra's Aquila neutral-atom processor, leveraging its programmable atom arrays to physically realize the reservoir dynamics.

quant-pharXiv:2512.18612•Dec 23, 2025•3 citations

Parameter-Efficient Multilinear Readout for Quantum Reservoir Computing

Van Tien Nguyen, Panagiotis Markopoulos

Quantum Reservoir Computing (QRC) uses fixed quantum dynamics as a high-dimensional feature map. Still, standard pipelines flatten reservoir measurements and train a linear readout, ignoring the natural tensor structure across time steps, qubits, and observables. We propose a multilinear readout for quantum reservoir computing (ML-QRC), a tensor-structured neural readout compressed through mode-wise projections. Relative to a dense hidden-width-64 neural readout, this reduces the feature-to-hidden parameter count from $64TnF$ to $d(T+n+F)+64d^3$, giving a $12.1\times$ feature-block compression at $T=50$, $n=6$, $d=8$, and $F=n+\binom{n}{2}$. Across Mackey-Glass forecasting and TFIM phase classification, ML-QRC remains statistically competitive with Ridge on Mackey-Glass and obtains the numerically highest full-data TFIM accuracy among tested methods, while using $12.0\times$ fewer parameters than the full dense MLP. Its clearest benefit is improved low-data stability relative to neural MLP readouts on Mackey-Glass; this advantage does not transfer to TFIM, where Ridge performs best at 50\% data. Overall, ML-QRC is best viewed as a tensor-structured neural readout that compresses dense MLP readouts while preserving QRC tensor structure, rather than as a replacement for simple linear/Ridge readouts in the tested regime.

quant-pharXiv:2610.13974•Sep 21, 2026•0 citations
We propose a hybrid photonic quantum reservoir computing (QRC) framework for swaption surface prediction. The pipeline compresses 224-dimensional surfaces to a 20-dimensional latent space via a sparse denoising autoencoder, extracts 1,215 Fock-basis features from an ensemble of three fixed photonic reservoirs, concatenates them with a 120-dimensional classical context, and maps the resulting 1,335-dimensional feature vector to predictions with Ridge regression. We benchmark against 10 classical and quantum baselines on six held-out trading days. Our approach achieves the lowest surface RMSE of~0.04250.04250.0425 while maintaining sub-millisecond inference. The quantum layer has zero trainable parameters, sidestepping barren plateaus entirely. Variational quantum methods (VQC, Quantum LSTM) yield negative R2R^{2}R2 on test data, confirming that fixed quantum feature extractors paired with regularised readouts are more viable for low-data financial applications.
Abstract Quantum machine learning is increasingly proposed for financial time series, yet practitioners have no way to predict whether quantum reservoir computing (QRC) will outperform classical baselines on a given dataset before committing resources. We propose a pre-experimental screening framework, built on persistence R 2 and principal component analysis (PCA) variance decomposition, that makes this prediction possible. We validate the framework using a 10-mode Boson sampling circuit (252 features) on three market datasets (US Treasury yields, CBOE VIX, crude oil) and a diagnostic nonlinear control. On the market data, the screening tool correctly predicts a structural ceiling : because financial surfaces follow near-random-walk dynamics (persistence R 2 > 0.95 , 3 PCA components), QRC collapses to the performance of classical Ridge regression ( R 2 = 0.9797 vs 0.9798 on Treasuries, with persistence at 0.9793 ). On the nonlinear control, the framework correctly predicts that a nonlinear feature map will outperform linear Ridge. Running the actual Boson sampling QRC on this control confirms the prediction: the quantum circuit achieves R 2 = 0.816 ± 0.007 , cleanly outperforming classical Ridge ( R 2 = 0.77 ) on a dataset where persistence completely fails ( R 2 = − 1.64 ). The proposed screening test can be computed from the raw data alone, before any quantum circuit is built.
This study explores quantum and classical hybrid architectures for financial time-series fore casting, focusing on Quantum Long Short-Term Memory (QLSTM) networks and Quantum Reservoir Computing (QRC), using univariate and multivariate lag structures on real financial data. We assess how lag embeddings affect predictive accuracy and robustness. Data are en coded into quantum states via amplitude encoding, enabling efficient representation of normalized lagged observations under realistic qubit constraints. The recurrent dynamics of QLSTM and the reservoir of QRC are implemented as parameterized quantum circuits, while classical optimizers train the readout and, where applicable, variational circuit parameters. We benchmark quantum models against classical LSTM and reservoir computing using common error like metrics. Our results show that, with suitable lag selection and amplitude encoding, quantum-enhanced archi tectures match classical baselines in univariate settings and can modestly outperform them in multivariate regimes with correlated inputs, where expressive encodings are most beneficial.
This study evaluates quantum classical hybrid recurrent architectures for financial time-series forecasting: a Quantum Long Short-Term Memory (QLSTM) network and Quantum Reservoir Computing (QRC) with lag-based inputs. Classical lag vectors are encoded into quantum states via amplitude encoding, which compactly represents normalized observations but requires nontrivial state preparation; the QLSTM gates and the QRC reservoir are implemented with parameterized quantum circuits, while classical optimizers train the interface and readout layers. On a monthly revenue series extended to 1056 months by a Gaussian-process--hidden-Markov generator, with a fixed chronological 70/15/15 split and one-step-ahead forecasts, we compare both models against a full-size LSTM, a classical control matched in parameter budget and interface architecture, a small plain LSTM, and classical reservoir computing, over ten seeds and against persistence, seasonal-naive and SARIMA benchmark forecasts. The QLSTM, with 2,105 parameters, is statistically indistinguishable from an LSTM 126 times larger, but equally indistinguishable from the matched classical control and from a plain LSTM of the same budget, so its parameter efficiency cannot be attributed to the quantum circuits. The quantum reservoir is worse than a classical reservoir of equal feature dimension in all ten seeds and worse than SARIMA, whereas all recurrent models and both classical reservoirs improve on SARIMA by about 2\% in RMSE. In a 24-month rolling-origin backtest the classical LSTMs degrade more slowly than the QLSTM. We find no evidence of a quantum accuracy or parameter-efficiency advantage in this low-dimensional setting.
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.
Quantum Reservoir Computing (QRC) leverages the natural dynamics of quantum systems for information processing, without requiring a fault-tolerant quantum computer. In this work, we apply QRC within a hybrid quantum classical framework for image denoising. The quantum reservoir is implemented using a Rydberg atom array, while a classical neural network serves as the readout layer. To prepare the input, images are first compressed using Principal Component Analysis (PCA), reducing their dimensionality to match the size of the atom array. Each feature vector is encoded into local detuning parameters of a time-dependent Hamiltonian governing the Rydberg system. As the system evolves, it generates nonlinear embeddings through the measurement of observables across multiple time steps. These temporal embeddings capture complex correlations, which are fed into a classical neural network to reconstruct the denoised images. To evaluate performance, we compare this QRC-assisted model against a baseline architecture consisting of PCA followed by a dense neural network, trained under identical conditions. Our results show that the QRC-based approach achieves improved image sharpness and similar structural recovery compared to the PCA-based model. We demonstrate the practical viability of this framework through experiments on QuEra's Aquila neutral-atom processor, leveraging its programmable atom arrays to physically realize the reservoir dynamics.
Quantum Reservoir Computing (QRC) uses fixed quantum dynamics as a high-dimensional feature map. Still, standard pipelines flatten reservoir measurements and train a linear readout, ignoring the natural tensor structure across time steps, qubits, and observables. We propose a multilinear readout for quantum reservoir computing (ML-QRC), a tensor-structured neural readout compressed through mode-wise projections. Relative to a dense hidden-width-64 neural readout, this reduces the feature-to-hidden parameter count from 64TnF64TnF64TnF to d(T+n+F)+64d3d(T+n+F)+64d^3d(T+n+F)+64d3, giving a 12.1×12.1\times12.1× feature-block compression at T=50T=50T=50, n=6n=6n=6, d=8d=8d=8, and F=n+(n2)F=n+\binom{n}{2}F=n+(2n​). Across Mackey-Glass forecasting and TFIM phase classification, ML-QRC remains statistically competitive with Ridge on Mackey-Glass and obtains the numerically highest full-data TFIM accuracy among tested methods, while using 12.0×12.0\times12.0× fewer parameters than the full dense MLP. Its clearest benefit is improved low-data stability relative to neural MLP readouts on Mackey-Glass; this advantage does not transfer to TFIM, where Ridge performs best at 50\% data. Overall, ML-QRC is best viewed as a tensor-structured neural readout that compresses dense MLP readouts while preserving QRC tensor structure, rather than as a replacement for simple linear/Ridge readouts in the tested regime.