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:2601.04592

Density Matrix RNN (DM-RNN): A Quantum Information Theoretic Framework for Modeling Musical Context and Polyphony

Joonwon Seo, Mariana Montiel

Published Jan 8, 2026• 0 citations
arXivHTMLPDF

Abstract

Classical Recurrent Neural Networks (RNNs) summarize musical context into a deterministic hidden state vector, imposing an information bottleneck that fails to capture the inherent ambiguity in music. We propose the Density Matrix RNN (DM-RNN), a novel theoretical architecture utilizing the Density Matrix. This allows the model to maintain a statistical ensemble of musical interpretations (a mixed state), capturing both classical probabilities and quantum coherences. We rigorously define the temporal dynamics using Quantum Channels (CPTP maps). Crucially, we detail a parameterization strategy based on the Choi-Jamiolkowski isomorphism, ensuring the learned dynamics remain physically valid (CPTP) by construction. We introduce an analytical framework using Von Neumann Entropy to quantify musical uncertainty and Quantum Mutual Information (QMI) to measure entanglement between voices. The DM-RNN provides a mathematically rigorous framework for modeling complex, ambiguous musical structures.

DOI
10.48550/arxiv.2601.04592

Related papers

Recurrent convolutional neural networks for modeling nonadiabatic dynamics of quantum-classical systems

Alex Ning, Lingyu Yang, Gia-Wei Chern

Recurrent neural networks (RNNs) have recently been extensively applied to model the time evolution in fluid dynamics, weather predictions, and even chaotic systems due to their ability to capture temporal dependencies and sequential patterns in data. Here we present an RNN model based on convolutional neural networks for modeling the nonlinear nonadiabatic dynamics of hybrid quantum-classical systems. The dynamical evolution of the hybrid systems is governed by equations of motion for classical degrees of freedom and von Neumann equation for electrons. The Physics-Aware Recurrent Convolution (PARC) neural network structure incorporates a differentiator-integrator architecture that inductively models the spatiotemporal dynamics of generic physical systems. We apply our RNN approach to learn the space-time evolution of a one-dimensional semiclassical Holstein model after an interaction quench. For shallow quenches (small changes in electron-lattice coupling), the deterministic dynamics can be accurately captured using a single-CNN-based recurrent network. In contrast, deep quenches induce chaotic evolution, making long-term trajectory prediction significantly more challenging. Nonetheless, we demonstrate that the PARC-CNN architecture can effectively learn the statistical climate of the Holstein model under deep-quench conditions.

arXiv:2412.06631•Physical review. E•Dec 17, 2025•0 citations

Detecting Markovianity of quantum processes via recurrent neural networks

Angela Rosy Morgillo, Massimiliano Federico Sacchi, Chiara Macchiavello

Abstract We present a novel methodology utilizing recurrent neural networks (RNNs) to classify Markovian and non-Markovian quantum processes, leveraging time series data derived from Choi states. The model exhibits exceptional accuracy, surpassing 95%, across diverse scenarios, including dephasing and Pauli channels in an arbitrary basis, generalized amplitude damping dynamics, and even in the presence of noise. Additionally, the developed model shows efficient forecasting capabilities for the analyzed time series data. These results suggest the potential of RNNs in discerning and predicting the Markovian nature of quantum processes.

arXiv:2406.07226•Quantum Machine Intelligence•Mar 27, 2025•1 citation

Quantum RNNs and LSTMs through Entangling and Disentangling Power of Unitary Transformations

Ammar Daskin

In this paper, we present a framework for modeling quantum recurrent neural networks (RNNs) and their enhanced version, long short-term memory (LSTM) networks using the core ideas presented by Linden et al. (2009), where the entangling and disentangling power of unitary transformations is investigated. In particular, we interpret entangling and disentangling power as information retention and forgetting mechanisms in LSTMs. Thus, entanglement emerges as a key component of the optimization (training) process. We believe that, by leveraging prior knowledge of the entangling power of unitaries, the proposed quantum-classical framework can guide the design of better-parameterized quantum circuits for various real-world applications.

arXiv:2505.06774•International Conference on Agents and Artificial Intelligence (ICAART)•Jan 1, 2026•0 citations

Hybrid Quantum-Classical Recurrent Neural Networks

Wenduan Xu

We present a hybrid quantum-classical recurrent neural network (QRNN) architecture in which the recurrent core is realized as a parametrized quantum circuit (PQC) controlled by a classical feedforward network. The hidden state is the quantum state of an $n$-qubit PQC in an exponentially large Hilbert space $\mathbb{C}^{2^n}$, which serves as a coherent recurrent quantum memory. The PQC is unitary by construction, making the hidden-state evolution norm-preserving without external constraints. At each timestep, mid-circuit Pauli expectation-value readouts are combined with the input embedding and processed by the feedforward network, which provides explicit classical nonlinearity. The outputs parametrize the PQC, which updates the hidden state via unitary dynamics. The QRNN is compact and physically consistent, and it unifies (i) unitary recurrence as a high-capacity memory, (ii) partial observation via mid-circuit readouts, and (iii) nonlinear classical control for input-conditioned parametrization. We evaluate the model in simulation with up to 14 qubits on sentiment analysis, MNIST, permuted MNIST, copying memory, and language modeling. For sequence-to-sequence learning, we further devise a soft attention mechanism over the mid-circuit readouts and show its effectiveness for machine translation. To our knowledge, this is the first model (RNN or otherwise) grounded in quantum operations to achieve competitive performance against strong classical baselines across a broad class of sequence-learning tasks.

arXiv:2510.25557•Oct 29, 2025•0 citations

Quantum-Like Representation of Dynamic Functional Connectivity: Double Covariance Formalism

Andrei Yu. Khrennikov, Makio Yamada, Irina Basieva

We introduce a neurophysiological adaptation of the Double Covariance Model (DCM) to provide a generative, fourth-order statistical framework for brain dynamics. Moving beyond descriptive sliding-window methods, the proposed theory implements a two-scale temporal scheme that explicitly separates fast micro-time stochastic neural fluctuations from the macro-time scale of emergent cognitive states. By treating the variance of covariance as a first-class mathematical object, the model computes the fourth-order moment of localized micro-signals to directly reconstruct a complex-valued network density matrix ($\rho$). This architecture transitions classical neurophysiological data into a quantum-like state representation, enabling the direct application of quantum information measures—such as concurrence and von Neumann entropy—to model macroscopic integration. Ultimately, this framework establishes a principled, non-phenomenological link between low-level stochastic network dynamics and the unified macroscopic phenomenon of "mental entanglement".

•Jun 23, 2026•0 citations

Quantum neural networks for machine learning applied to the tracking and control of the dynamics of stochastic transmission lines

Harish Parthasarathy, Arti Vaish

In most of the machine learning problems based on classical neural networks, we try to identify a nonlinear system which may or may not have memory by adapting the weights of the neural network so that for the same input process applied to the system as well as to the neural network, the network output matches that of the true system output in the mean square sense or in the sense of some other metric on the space of output processes. If the network is feed forward, then this model can be used to simulate a memory-less nonlinear system. If however the actual system has memory and is described by a stochastic difference equation for its state followed by an output equation that is a function of the state, then we can use a recurrent neural network (RNN) whose state dynamics are described by a nonlinear difference equation involving weights as parameters and the state and weights of the RNN can be estimated from noisy measurements of the actual system output by means of the extended Kalman filter. If however, we are interested in estimating the joint probability density of a vector valued process like the transmission line current and voltage at different points along the line at all times, then a classical neural network will not work. This is because a probability density has several constraints like non-negativity and integrability to unity, and so on, and a classical neural network does not take care of these constraints.

CRC Press eBooks•Aug 18, 2023•1 citation
Classical Recurrent Neural Networks (RNNs) summarize musical context into a deterministic hidden state vector, imposing an information bottleneck that fails to capture the inherent ambiguity in music. We propose the Density Matrix RNN (DM-RNN), a novel theoretical architecture utilizing the Density Matrix. This allows the model to maintain a statistical ensemble of musical interpretations (a mixed state), capturing both classical probabilities and quantum coherences. We rigorously define the temporal dynamics using Quantum Channels (CPTP maps). Crucially, we detail a parameterization strategy based on the Choi-Jamiolkowski isomorphism, ensuring the learned dynamics remain physically valid (CPTP) by construction. We introduce an analytical framework using Von Neumann Entropy to quantify musical uncertainty and Quantum Mutual Information (QMI) to measure entanglement between voices. The DM-RNN provides a mathematically rigorous framework for modeling complex, ambiguous musical structures.
Recurrent neural networks (RNNs) have recently been extensively applied to model the time evolution in fluid dynamics, weather predictions, and even chaotic systems due to their ability to capture temporal dependencies and sequential patterns in data. Here we present an RNN model based on convolutional neural networks for modeling the nonlinear nonadiabatic dynamics of hybrid quantum-classical systems. The dynamical evolution of the hybrid systems is governed by equations of motion for classical degrees of freedom and von Neumann equation for electrons. The Physics-Aware Recurrent Convolution (PARC) neural network structure incorporates a differentiator-integrator architecture that inductively models the spatiotemporal dynamics of generic physical systems. We apply our RNN approach to learn the space-time evolution of a one-dimensional semiclassical Holstein model after an interaction quench. For shallow quenches (small changes in electron-lattice coupling), the deterministic dynamics can be accurately captured using a single-CNN-based recurrent network. In contrast, deep quenches induce chaotic evolution, making long-term trajectory prediction significantly more challenging. Nonetheless, we demonstrate that the PARC-CNN architecture can effectively learn the statistical climate of the Holstein model under deep-quench conditions.
Abstract We present a novel methodology utilizing recurrent neural networks (RNNs) to classify Markovian and non-Markovian quantum processes, leveraging time series data derived from Choi states. The model exhibits exceptional accuracy, surpassing 95%, across diverse scenarios, including dephasing and Pauli channels in an arbitrary basis, generalized amplitude damping dynamics, and even in the presence of noise. Additionally, the developed model shows efficient forecasting capabilities for the analyzed time series data. These results suggest the potential of RNNs in discerning and predicting the Markovian nature of quantum processes.
In this paper, we present a framework for modeling quantum recurrent neural networks (RNNs) and their enhanced version, long short-term memory (LSTM) networks using the core ideas presented by Linden et al. (2009), where the entangling and disentangling power of unitary transformations is investigated. In particular, we interpret entangling and disentangling power as information retention and forgetting mechanisms in LSTMs. Thus, entanglement emerges as a key component of the optimization (training) process. We believe that, by leveraging prior knowledge of the entangling power of unitaries, the proposed quantum-classical framework can guide the design of better-parameterized quantum circuits for various real-world applications.
We present a hybrid quantum-classical recurrent neural network (QRNN) architecture in which the recurrent core is realized as a parametrized quantum circuit (PQC) controlled by a classical feedforward network. The hidden state is the quantum state of an nnn-qubit PQC in an exponentially large Hilbert space C2n\mathbb{C}^{2^n}C2n, which serves as a coherent recurrent quantum memory. The PQC is unitary by construction, making the hidden-state evolution norm-preserving without external constraints. At each timestep, mid-circuit Pauli expectation-value readouts are combined with the input embedding and processed by the feedforward network, which provides explicit classical nonlinearity. The outputs parametrize the PQC, which updates the hidden state via unitary dynamics. The QRNN is compact and physically consistent, and it unifies (i) unitary recurrence as a high-capacity memory, (ii) partial observation via mid-circuit readouts, and (iii) nonlinear classical control for input-conditioned parametrization. We evaluate the model in simulation with up to 14 qubits on sentiment analysis, MNIST, permuted MNIST, copying memory, and language modeling. For sequence-to-sequence learning, we further devise a soft attention mechanism over the mid-circuit readouts and show its effectiveness for machine translation. To our knowledge, this is the first model (RNN or otherwise) grounded in quantum operations to achieve competitive performance against strong classical baselines across a broad class of sequence-learning tasks.
We introduce a neurophysiological adaptation of the Double Covariance Model (DCM) to provide a generative, fourth-order statistical framework for brain dynamics. Moving beyond descriptive sliding-window methods, the proposed theory implements a two-scale temporal scheme that explicitly separates fast micro-time stochastic neural fluctuations from the macro-time scale of emergent cognitive states. By treating the variance of covariance as a first-class mathematical object, the model computes the fourth-order moment of localized micro-signals to directly reconstruct a complex-valued network density matrix (ρ\rhoρ). This architecture transitions classical neurophysiological data into a quantum-like state representation, enabling the direct application of quantum information measures—such as concurrence and von Neumann entropy—to model macroscopic integration. Ultimately, this framework establishes a principled, non-phenomenological link between low-level stochastic network dynamics and the unified macroscopic phenomenon of "mental entanglement".
In most of the machine learning problems based on classical neural networks, we try to identify a nonlinear system which may or may not have memory by adapting the weights of the neural network so that for the same input process applied to the system as well as to the neural network, the network output matches that of the true system output in the mean square sense or in the sense of some other metric on the space of output processes. If the network is feed forward, then this model can be used to simulate a memory-less nonlinear system. If however the actual system has memory and is described by a stochastic difference equation for its state followed by an output equation that is a function of the state, then we can use a recurrent neural network (RNN) whose state dynamics are described by a nonlinear difference equation involving weights as parameters and the state and weights of the RNN can be estimated from noisy measurements of the actual system output by means of the extended Kalman filter. If however, we are interested in estimating the joint probability density of a vector valued process like the transmission line current and voltage at different points along the line at all times, then a classical neural network will not work. This is because a probability density has several constraints like non-negativity and integrability to unity, and so on, and a classical neural network does not take care of these constraints.