Abstract Entangled quantum states are highly sensitive to noise, which makes it difficult to transfer them over noisy quantum channels or to store them in quantum memory. Here, we propose the disentangling quantum autoencoder (DQAE) to encode entangled states into single-qubit product states. The DQAE provides an exponential improvement in the number of copies needed to transport entangled states across qubit-loss or leakage channels compared to unencoded states. The DQAE can be trained in an unsupervised manner from entangled quantum data. For general states, we train via variational quantum algorithms based on gradient descent with purity-based cost functions, while stabilizer states can be trained via a Metropolis algorithm. For particular classes of states, the number of training data needed to generalize is surprisingly low: for stabilizer states, DQAE generalizes by learning from a number of training data that scales linearly with the number of qubits, while only 1 training sample is sufficient for states evolved with the transverse-field Ising Hamiltonian. Our work provides practical applications for enhancing near-term quantum computers.
Abstract Entangled quantum states are highly sensitive to noise, which makes it difficult to transfer them over noisy quantum channels or to store them in quantum memory. Here, we propose the disentangling quantum autoencoder (DQAE) to encode entangled states into single-qubit product states. The DQAE provides an exponential improvement in the number of copies needed to transport entangled states across qubit-loss or leakage channels compared to unencoded states. The DQAE can be trained in an unsupervised manner from entangled quantum data. For general states, we train via variational quantum algorithms based on gradient descent with purity-based cost functions, while stabilizer states can be trained via a Metropolis algorithm. For particular classes of states, the number of training data needed to generalize is surprisingly low: for stabilizer states, DQAE generalizes by learning from a number of training data that scales linearly with the number of qubits, while only 1 training sample is sufficient for states evolved with the transverse-field Ising Hamiltonian. Our work provides practical applications for enhancing near-term quantum computers.
Abstract Quantum machine learning methods often rely on fixed, hand-crafted quantum encodings that may not capture optimal features for downstream tasks. In this work, we study the power of quantum autoencoders in learning data-driven quantum representations. We first theoretically demonstrate that the quantum autoencoder method is efficient in terms of sample complexity throughout the entire training process. Then we numerically train the quantum autoencoder on 3 million peptide sequences, and evaluate their effectiveness across multiple peptide classification problems, including antihypertensive activity, blood–brain barrier-penetration, and cytotoxicity. The learned representations were compared against Hamiltonian-evolved baselines using a quantum kernel with support vector machines. Results show that quantum autoencoder learned representations achieve accuracy improvements ranging from 0.4% to 8.1% over Hamiltonian baselines across seven datasets, demonstrating effective generalization across biologically distinct datasets, with pre-training enabling effective transfer learning without task-specific fine-tuning. This work establishes that quantum autoencoder architectures can effectively learn from large-scale datasets (3 million samples) with compact parameterizations (~900 parameters), demonstrating their viability for practical quantum applications.
Variational autoencoders are powerful representation learning models that map complex data into low-dimensional latent spaces, enabling the discovery of interpretable and disentangled factors. Such representations can facilitate the interpretation and controllable generation of data describing complex scientific systems. Understanding how these factors are organized and encoded in latent space is therefore important for developing reliable representation learning models. Recently, quantum variational autoencoders (QVAEs) have been proposed as quantum representation models, demonstrating informative latent representations and improved latent-space occupancy through quantum regularization. However, it remains unclear whether and how QVAEs can learn disentangled and interpretable latent factors. A key challenge in investigating quantum latent factors is that a small number of qubits spans an exponentially large Hilbert space, making the notion of an individual quantum latent dimension nontrivial. Here, we investigate what constitutes an individual quantum latent dimension and whether it can encode a distinct factor. We develop theoretical insights into quantum latent dimensions and support them with empirical studies on representative synthetic problems, including MNIST variants. Across three datasets, we demonstrate that QVAEs can discover factorized and semantically interpretable latent representations, with individual qubits functioning as meaningful latent factors. These results establish a foundation for understanding quantum latent spaces and their potential for structured and interpretable representation learning.
Quantum machine learning methods often rely on fixed, hand-crafted quantum encodings that may not capture optimal features for downstream tasks. In this work, we study the power of quantum autoencoders in learning data-driven quantum representations. We first theoretically demonstrate that the quantum autoencoder method is efficient in terms of sample complexity throughout the entire training process. Then we numerically train the quantum autoencoder on 3 million peptide sequences, and evaluate their effectiveness across multiple peptide classification problems including antihypertensive peptide prediction, blood-brain barrier-penetration, and cytotoxic activity detection. The learned representations were compared against Hamiltonian-evolved baselines using a quantum kernel with support vector machines. Results show that quantum autoencoder learned representations achieve accuracy improvements ranging from 0.4\% to 8.1\% over Hamiltonian baselines across seven datasets, demonstrating effective generalization to diverse downstream datasets with pre-training enabling effective transfer learning without task-specific fine-tuning. This work establishes that quantum autoencoder architectures can effectively learn from large-scale datasets (3 million samples) with compact parameterizations ($\sim$900 parameters), demonstrating their viability for practical quantum applications.
Quantum state classification and entanglement quantification are of significant importance in the fundamental research of quantum information science and various quantum applications. Traditional methods, such as quantum state tomography, face exponential measurement demands with increasing numbers of qubits, necessitating more efficient approaches. Recent work has shown promise in using artificial neural networks (ANNs) for quantum state analysis. However, existing ANNs may falter when confronted with states affected by dephasing noise, especially with limited data and computational resources. In this study, we employ a machine-learning approach to investigate the disentanglement process in two-qubit systems in the presence of dephasing noise. Our findings highlight the limitations of general state-trained ANNs in classifying states under dephasing noise. Specialized ANN algorithms, tailored for classifying states and quantifying entanglement in such noisy environments, demonstrate excellent performance using only a subset of tomographic features.
An important task in quantum generative machine learning is to model the probability distribution of measurements of many-body quantum systems. Classical generative models, such as generative adversarial networks (GANs) and variational autoencoders (VAEs), can model the distributions of product states with high fidelity, but fail or require an exponential number of parameters to model entangled states. In this paper, we introduce a quantum-enhanced VAE (QeVAE), a generative quantum-classical hybrid model that uses quantum correlations to improve the fidelity over classical VAEs, while requiring only a linear number of parameters. We provide a closed-form expression for the output distributions of the QeVAE. We also empirically show that the QeVAE outperforms classical models on several classes of quantum states, such as 4-qubit and 8-qubit quantum circuit states, haar random states, and quantum kicked rotor states, with a more than 2x increase in fidelity for some states. Finally, we find that the trained model outperforms the classical model when executed on the IBMq Manila quantum computer. Our work paves the way for new applications of quantum generative learning algorithms and characterizing measurement distributions of high-dimensional quantum states.
Quantum autoencoder is an efficient variational quantum algorithm for quantum data compression. However, previous quantum autoencoders fail to compress and recover high-rank mixed states. In this work, we discuss the fundamental properties and limitations of the standard quantum-autoencoder model in more depth, and provide an information-theoretic solution to its recovering fidelity. Based on this understanding, we present a noise-assisted quantum-autoencoder algorithm to go beyond the limitations, our model can achieve high recovering fidelity for general input states. Appropriate noise channels are used to make the input mixedness and output mixedness consistent, the noise setup is determined by measurement results of the trash system. Compared with the original quantum-autoencoder model, the measurement information is fully used in our algorithm. In addition to the circuit model, we design a (noise-assisted) adiabatic model of quantum autoencoder that can be implemented on quantum annealers. We verify the validity of our methods through compressing the thermal states of the transverse-field Ising model and Werner states. For pure-state ensemble compression, we also introduce a projected quantum-autoencoder algorithm.