{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T16:13:42Z","timestamp":1778256822164,"version":"3.51.4"},"reference-count":60,"publisher":"IOP Publishing","issue":"1","license":[{"start":{"date-parts":[[2025,1,30]],"date-time":"2025-01-30T00:00:00Z","timestamp":1738195200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,1,30]],"date-time":"2025-01-30T00:00:00Z","timestamp":1738195200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/501100010769","name":"Axencia Galega de Innovaci\u00f3n","doi-asserted-by":"crossref","award":["IN606A-2023\/011"],"award-info":[{"award-number":["IN606A-2023\/011"]}],"id":[{"id":"10.13039\/501100010769","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100010801","name":"Xunta de Galicia","doi-asserted-by":"crossref","award":["PRTR-C17.I1"],"award-info":[{"award-number":["PRTR-C17.I1"]}],"id":[{"id":"10.13039\/501100010801","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Ministerio de Ciencia, Innovaci\u00f3n y Universidades","award":["PRTR-C17.I1"],"award-info":[{"award-number":["PRTR-C17.I1"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,3,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Quantum recurrent neural networks (QRNNs) are robust candidates for modelling and predicting future values in multivariate time series. However, the effective implementation of some QRNN models is limited by the need for mid-circuit measurements. Those increase the requirements for quantum hardware, which in the current noisy intermediate-scale quantum era does not allow reliable computations. Emulation arises as the main near-term alternative to explore the potential of QRNNs, but existing quantum emulators are not dedicated to circuits with multiple intermediate measurements. In this context, we design a specific emulation method that relies on density matrix formalism. Using a compact tensor notation, we provide the mathematical formulation of the operator-sum representation involved. This allows us to show how the present and past information from a time series is transmitted through the circuit, and how to reduce the computational cost in every time step of the emulated network. In addition, we derive the analytical gradient and the Hessian of the network outputs with respect to its trainable parameters, which are needed when the outputs have stochastic noise due to hardware errors and a finite number of circuit shots (sampling). We finally test the presented methods using a hardware-efficient ansatz and four diverse datasets that include univariate and multivariate time series, with and without sampling noise. In addition, we compare the model with other existing quantum and classical approaches. Our results show how QRNNs can be trained with numerical and analytical gradients to make accurate predictions of future values by capturing non-trivial patterns of input series with different complexities.<\/jats:p>","DOI":"10.1088\/2632-2153\/ad9431","type":"journal-article","created":{"date-parts":[[2024,11,18]],"date-time":"2024-11-18T23:01:11Z","timestamp":1731970871000},"page":"015023","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":8,"title":["Density matrix emulation of quantum recurrent neural networks for multivariate time series prediction"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0041-6802","authenticated-orcid":true,"given":"J D","family":"Viqueira","sequence":"first","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7685-4331","authenticated-orcid":false,"given":"D","family":"Fa\u00edlde","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0009-0005-0692-7098","authenticated-orcid":false,"given":"M M","family":"Juane","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7272-8488","authenticated-orcid":false,"given":"A","family":"G\u00f3mez","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0639-6574","authenticated-orcid":false,"given":"D","family":"Mera","sequence":"additional","affiliation":[]}],"member":"266","published-online":{"date-parts":[[2025,1,30]]},"reference":[{"key":"mlstad9431bib1","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"key":"mlstad9431bib2","doi-asserted-by":"publisher","first-page":"35","DOI":"10.1016\/S0169-2070(97)00044-7","article-title":"Forecasting with artificial neural networks: the state of the art","volume":"14","author":"Zhang","year":"1998","journal-title":"Int. J. Forecast."},{"key":"mlstad9431bib3","doi-asserted-by":"publisher","first-page":"159","DOI":"10.1016\/S0925-2312(01)00702-0","article-title":"Time series forecasting using a hybrid ARIMA and neural network model","volume":"50","author":"Zhang","year":"2003","journal-title":"Neurocomputing"},{"key":"mlstad9431bib4","doi-asserted-by":"publisher","first-page":"1724","DOI":"10.3115\/v1\/D14-1179","article-title":"Learning phrase representations using RNN encoder-decoder for statistical machine translation","author":"Cho","year":"2014"},{"key":"mlstad9431bib5","doi-asserted-by":"publisher","first-page":"95","DOI":"10.1145\/3209978.3210006","article-title":"Modeling long- and short-term temporal patterns with deep neural networks","author":"Lai","year":"2018"},{"key":"mlstad9431bib6","doi-asserted-by":"publisher","first-page":"1409","DOI":"10.1609\/aaai.v33i01.33011409","article-title":"A deep neural network for unsupervised anomaly detection and diagnosis in multivariate time series data","volume":"vol 33","author":"Zhang","year":"2019"},{"key":"mlstad9431bib7","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3444690","article-title":"A review on outlier\/anomaly detection in time series data","volume":"54","author":"Bl\u00e1zquez-Garc\u00eda","year":"2021","journal-title":"ACM Comput. Surv."},{"key":"mlstad9431bib8","article-title":"A critical review of recurrent neural networks for sequence learning","author":"Lipton","year":"2015"},{"key":"mlstad9431bib9","doi-asserted-by":"publisher","first-page":"103","DOI":"10.3115\/v1\/W14-4012","article-title":"On the properties of neural machine translation: encoder-decoder approaches","author":"Cho","year":"2014"},{"key":"mlstad9431bib10","article-title":"Attention is all you need","volume":"vol 30","author":"Vaswani","year":"2017"},{"key":"mlstad9431bib11","doi-asserted-by":"publisher","first-page":"85","DOI":"10.1016\/j.neunet.2014.09.003","article-title":"Deep learning in neural networks: an overview","volume":"61","author":"Schmidhuber","year":"2015","journal-title":"Neural Netw."},{"key":"mlstad9431bib12","doi-asserted-by":"publisher","first-page":"32","DOI":"10.22331\/qv-2020-03-17-32","article-title":"A non-review of quantum machine learning: trends and explorations","volume":"4","author":"Dunjko","year":"2020","journal-title":"Quantum Views"},{"key":"mlstad9431bib13","doi-asserted-by":"publisher","first-page":"195","DOI":"10.1038\/nature23474","article-title":"Quantum machine learning","volume":"549","author":"Biamonte","year":"2017","journal-title":"Nature"},{"key":"mlstad9431bib14","doi-asserted-by":"publisher","first-page":"625","DOI":"10.1038\/s42254-021-00348-9","article-title":"Variational quantum algorithms","volume":"3","author":"Cerezo","year":"2021","journal-title":"Nat. Rev. Phys."},{"key":"mlstad9431bib15","doi-asserted-by":"publisher","first-page":"309","DOI":"10.1103\/PhysRevA.98.032309","article-title":"Quantum circuit learning","volume":"98","author":"Mitarai","year":"2018","journal-title":"Phys. Rev. A"},{"key":"mlstad9431bib16","doi-asserted-by":"publisher","first-page":"75","DOI":"10.1007\/3-540-49208-9_5","article-title":"Quantum recurrent networks for simulating stochastic processes","author":"Zak","year":"1999"},{"key":"mlstad9431bib17","doi-asserted-by":"publisher","first-page":"165","DOI":"10.1109\/ICISIP.2004.1287645","article-title":"Stochastic filtering and speech enhancement using a recurrent quantum neural network","author":"Behera","year":"2004"},{"key":"mlstad9431bib18","doi-asserted-by":"publisher","first-page":"357","DOI":"10.1007\/s10702-005-7125-6","article-title":"A recurrent quantum neural network model to describe eye tracking of moving targets","volume":"18","author":"Behera","year":"2005","journal-title":"Found. Phys. Lett."},{"key":"mlstad9431bib19","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2020.105863","article-title":"Quantum recurrent encoder-decoder neural network for performance trend prediction of rotating machinery","volume":"197","author":"Chen","year":"2020","journal-title":"Knowl.-Based Syst."},{"key":"mlstad9431bib20","doi-asserted-by":"publisher","first-page":"8622","DOI":"10.1109\/ICASSP43922.2022.9747369","article-title":"Quantum long short-term memory","author":"Chen","year":"2022"},{"key":"mlstad9431bib21","doi-asserted-by":"publisher","first-page":"9221","DOI":"10.1007\/s00521-020-05685-x","article-title":"Full-state neural network observer-based hybrid quantum diagonal recurrent neural network adaptive tracking control","volume":"33","author":"Elkenawy","year":"2021","journal-title":"Neural Comput. Appl."},{"key":"mlstad9431bib22","first-page":"1","article-title":"Quantum deep recurrent reinforcement learning","author":"Chen","year":"2023"},{"key":"mlstad9431bib23","doi-asserted-by":"publisher","first-page":"7231","DOI":"10.1038\/s41467-022-34780-x","article-title":"Path sampling of recurrent neural networks by incorporating known physics","volume":"13","author":"Tsai","year":"2022","journal-title":"Nat. Commun."},{"key":"mlstad9431bib24","first-page":"1368","article-title":"Recurrent quantum neural networks","volume":"vol 33","author":"Bausch","year":"2020"},{"key":"mlstad9431bib25","doi-asserted-by":"publisher","first-page":"414","DOI":"10.1103\/PhysRevA.103.052414","article-title":"Learning temporal data with a variational quantum recurrent neural network","volume":"103","author":"Takaki","year":"2021","journal-title":"Phys. Rev. A"},{"key":"mlstad9431bib26","doi-asserted-by":"publisher","first-page":"31","DOI":"10.1007\/s42484-023-00117-0","article-title":"Rapid training of quantum recurrent neural networks","volume":"5","author":"Siemaszko","year":"2023","journal-title":"Quantum Mach. Intel."},{"key":"mlstad9431bib27","article-title":"Learning quantum processes with memory -quantum recurrent neural networks","author":"Bondarenko","year":"2023"},{"key":"mlstad9431bib28","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1016\/j.neunet.2023.07.003","article-title":"Quantum recurrent neural networks for sequential learning","volume":"166","author":"Li","year":"2023","journal-title":"Neural Netw."},{"key":"mlstad9431bib29","article-title":"Quantum-discrete-map-based recurrent neural networks","author":"Sun","year":"2023"},{"key":"mlstad9431bib30","doi-asserted-by":"publisher","first-page":"501","DOI":"10.1103\/physrevlett.127.100501","article-title":"Exploiting dynamic quantum circuits in aquantum algorithm with superconducting qubits","volume":"127","author":"C\u00f3rcoles","year":"2021","journal-title":"Phys. Rev. Lett."},{"key":"mlstad9431bib31","article-title":"Bringing the full power of dynamic circuits to Qiskit Runtime","author":"IBM","year":"2022"},{"key":"mlstad9431bib32","article-title":"Experimental demonstration of the advantage of adaptive quantum circuits","author":"Foss-Feig","year":"2023"},{"key":"mlstad9431bib33","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1103\/PhysRevA.99.032331","article-title":"Evaluating analytic gradients on quantum hardware","volume":"99","author":"Schuld","year":"2019","journal-title":"Phys. Rev. A"},{"key":"mlstad9431bib34","author":"G\u00e9ron","year":"2017","edition":"1st edn"},{"key":"mlstad9431bib35","doi-asserted-by":"publisher","DOI":"10.1063\/1.5089550","article-title":"A quantum engineer\u2019s guide to superconducting qubits","volume":"6","author":"Krantz","year":"2019","journal-title":"Appl. Phys. Rev."},{"key":"mlstad9431bib36","doi-asserted-by":"publisher","first-page":"179","DOI":"10.1207\/s15516709cog1402_1","article-title":"Finding structure in time","volume":"14","author":"Elman","year":"1990","journal-title":"Cogn. Sci."},{"key":"mlstad9431bib37","doi-asserted-by":"crossref","DOI":"10.26421\/QIC15.9-10-3","article-title":"Tensor networks and graphical calculus for open quantum systems","author":"Wood","year":"2015"},{"key":"mlstad9431bib38","doi-asserted-by":"publisher","first-page":"5595","DOI":"10.5555\/3122009.3242010","article-title":"Automatic differentiation in machine learning: a survey","volume":"18","author":"Baydin","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"mlstad9431bib39","doi-asserted-by":"publisher","first-page":"1550","DOI":"10.1109\/5.58337","article-title":"Backpropagation through time: what it does and how to do it","volume":"78","author":"Werbos","year":"1990","journal-title":"Proc. IEEE"},{"key":"mlstad9431bib40","article-title":"Tutorial on training recurrent neural networks, covering BPPT, RTRL, EKF and the echo state network approach","volume":"5","author":"Jaeger","year":"2002","journal-title":"GMD-Forsch. Inf."},{"key":"mlstad9431bib41","article-title":"Dive into deep learning","author":"Zhang","year":"2023"},{"key":"mlstad9431bib42","article-title":"Unbiasing truncated backpropagation through time","author":"Tallec","year":"2017"},{"key":"mlstad9431bib43","doi-asserted-by":"publisher","first-page":"502","DOI":"10.1103\/PhysRevLett.126.140502","article-title":"Low-depth gradient measurements can improve convergence in variational hybrid quantum-classical algorithms","volume":"126","author":"Harrow","year":"2021","journal-title":"Phys. Rev. Lett."},{"key":"mlstad9431bib44","doi-asserted-by":"publisher","first-page":"677","DOI":"10.22331\/q-2022-03-30-677","article-title":"General parameter-shift rules for quantum gradients","volume":"6","author":"Wierichs","year":"2022","journal-title":"Quantum"},{"key":"mlstad9431bib45","doi-asserted-by":"publisher","DOI":"10.1088\/2058-9565\/ab4eb5","article-title":"Parameterized quantum circuits as machine learning models","volume":"4","author":"Benedetti","year":"2019","journal-title":"Quantum Sci. Technol."},{"key":"mlstad9431bib46","doi-asserted-by":"publisher","first-page":"4887","DOI":"10.5555\/3157382.3157643","article-title":"Full-capacity unitary recurrent neural networks","author":"Wisdom","year":"2016"},{"key":"mlstad9431bib47","doi-asserted-by":"publisher","first-page":"4812","DOI":"10.1038\/s41467-018-07090-4","article-title":"Barren plateaus in quantum neural network training landscapes","volume":"9","author":"McClean","year":"2018","journal-title":"Nat. Commun."},{"key":"mlstad9431bib48","doi-asserted-by":"publisher","first-page":"1791","DOI":"10.1038\/s41467-021-21728-w","article-title":"Cost function dependent barren plateaus in shallow parametrized quantum circuits","volume":"12","author":"Cerezo","year":"2021","journal-title":"Nat. Commun."},{"key":"mlstad9431bib49","doi-asserted-by":"publisher","DOI":"10.1088\/2058-9565\/abdbc9","article-title":"Characterizing the loss landscape of variational quantum circuits","volume":"6","author":"Huembeli","year":"2021","journal-title":"Quantum Sci. Technol."},{"key":"mlstad9431bib50","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1103\/PhysRevResearch.1.013006","article-title":"Methodology for replacing indirect measurements with direct measurements","volume":"1","author":"Mitarai","year":"2019","journal-title":"Phys. Rev. Res."},{"key":"mlstad9431bib51","doi-asserted-by":"publisher","first-page":"430","DOI":"10.1103\/PhysRevA.103.032430","article-title":"Effect of data encoding on the expressive power of variational quantum-machine-learning models","volume":"103","author":"Schuld","year":"2021","journal-title":"Phys. Rev. A"},{"key":"mlstad9431bib52","doi-asserted-by":"publisher","first-page":"226","DOI":"10.22331\/q-2020-02-06-226","article-title":"Data re-uploading for a universal quantum classifier","volume":"4","author":"P\u00e9rez-Salinas","year":"2020","journal-title":"Quantum"},{"key":"mlstad9431bib53","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevA.107.062612","article-title":"Multidimensional Fourier series with quantum circuits","volume":"107","author":"Casas","year":"2023","journal-title":"Phys. Rev. A"},{"key":"mlstad9431bib54","doi-asserted-by":"publisher","first-page":"6354","DOI":"10.1103\/PhysRevA.40.6354","article-title":"Dimensions and entropies of chaotic intensity pulsations in a single-mode far-infrared NH3 laser","volume":"40","author":"H\u00fcbner","year":"1989","journal-title":"Phys. Rev. A"},{"key":"mlstad9431bib55","doi-asserted-by":"publisher","first-page":"1786","DOI":"10.1109\/ICNN.1993.298828","article-title":"Results of the time series prediction competition at the Santa Fe Institute","author":"Weigend","year":"1993"},{"key":"mlstad9431bib56","doi-asserted-by":"publisher","first-page":"16","DOI":"10.1038\/s41534-023-00682-z","article-title":"Time-series quantum reservoir computing with weak and projective measurements","volume":"9","author":"Mujal","year":"2023","journal-title":"npj Quantum Inf."},{"key":"mlstad9431bib57","article-title":"Pennylane: automatic differentiation of hybrid quantum-classical computations","author":"Bergholm","year":"2022"},{"key":"mlstad9431bib58","article-title":"Adam: a method for stochastic optimization","author":"Kingma","year":"2017"},{"key":"mlstad9431bib59","article-title":"Pytorch: an imperative style, high-performance deep learning library","author":"Paszke","year":"2019"},{"key":"mlstad9431bib60","article-title":"Dask: library for dynamic task scheduling","author":"Dask Development Team","year":"2016"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,1,30]],"date-time":"2025-01-30T11:38:02Z","timestamp":1738237082000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad9431"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,30]]},"references-count":60,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2025,1,30]]},"published-print":{"date-parts":[[2025,3,31]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ad9431","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,30]]},"assertion":[{"value":"Density matrix emulation of quantum recurrent neural networks for multivariate time series prediction","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2024-07-19","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-11-18","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-01-30","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}