Quantum Long Short-Term Memory

Chen, Samuel Yen-Chi; Yoo, Shinjae; Fang, Yao-Lung L.

doi:10.1109/icassp43922.2022.9747369

Cited by 55 publications

(39 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 one can see that the quantum circuit minimizes loss significantly faster in the first few epochs in comparison to the classical network. The loss curve is also more smooth in the quantum case, as was also noticed in [2]. For bigger networks (about 30 parameters), we observe that the loss at the end of the training is lower than for the smaller network, demonstrating that the networks predictions improve with increasing numbers of qmodes and parameters.…”

Section: Resultssupporting

confidence: 76%

See 1 more Smart Citation

Rapid training of quantum recurrent neural network

Michał¹,

McDermott²,

Buraczewski³

et al. 2022

Preprint

View full text Add to dashboard Cite

Time series prediction is the crucial task for many human activities e.g. weather forecasts or predicting stock prices. One solution to this problem is to use Recurrent Neural Networks (RNNs). Although they can yield accurate predictions, their learning process is slow and complex. Here we propose a Quantum Recurrent Neural Network (QRNN) to address these obstacles. The design of the network is based on the continuous-variable quantum computing paradigm. We demonstrate that the network is capable of learning time dependence of a few types of temporal data. Our numerical simulations show that the QRNN converges to optimal weights in fewer epochs than the classical network. Furthermore, for a small number of trainable parameters it can achieve lower loss than the latter.

show abstract

Section: Resultssupporting

confidence: 76%

“…Recent papers show benefits of using this approach for satellite image classification [6] or modeling joint probability distributions [8]. QML was also used to analyse time series [1,7,2]. In this paper we propose new approach to the task of time series prediction using Quantum Recurrent Neural Networks in continuous variable paradigm.…”

Section: Introductionmentioning

confidence: 99%

Rapid training of quantum recurrent neural network

Michał¹,

McDermott²,

Buraczewski³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…VQC Circuit Ansatz, G(θ g ). Mainstream QNNs [13,16,17] adopt a VQC ansatz constructed by parameterized singlequbit rotation gates followed by nearest-neighbor coupling of qubits using fixed two-qubit CNOT gates, as illustrated in Figure 1. Such a circuit ansatz has demonstrated superior expressive capability in various applications.…”

Section: Quantum Gans Backgroundmentioning

confidence: 99%

“…It is clear that the same fidelity and hence the same loss can be achieved with different parameter sets due to the periodic repeatability of the sine and cosine functions, resulting in distorted generated images (See Figure 2). Although input data can be mapped to the range [0,π] before applying the angle encoding [8,16], such unfaithful normalization prevents a QNN from learning an accurate model and producing high-quality output. We therefore conclude that the de facto angle encoding (even with normalization) fails to ensure highquality output in a fidelity based GAN framework.…”

Section: Preliminary Analysismentioning

confidence: 99%

IQGAN: Robust Quantum Generative Adversarial Network for Image Synthesis On NISQ Devices

Cheng¹,

Skipper²,

Swany³

et al. 2022

Preprint

View full text Add to dashboard Cite

In this work, we propose IQGAN, a quantum Generative Adversarial Network (GAN) framework for multiqubit image synthesis that can be efficiently implemented on Noisy Intermediate Scale Quantum (NISQ) devices. We investigate the reasons for the inferior generative performance of current quantum GANs in our preliminary study and conclude that an adjustable input encoder is the key to ensuring highquality data synthesis. We then propose the IQGAN architecture featuring a trainable multiqubit quantum encoder that effectively embeds classical data into quantum states. Furthermore, we propose a compact quantum generator that significantly reduces the design cost and circuit depth on NISQ devices. Experimental results on both IBM quantum processors and quantum simulators demonstrated that IQGAN outperforms state-of-the-art quantum GANs in qualitative and quantitative evaluation of the generated samples, model convergence, and quantum computing cost.

show abstract

“…It is a model that is capable of inferencing fully on quantum computers. On the contrary, Quantum Long short-term memory (Chen, Yoo, and Fang 2022) was also proposed. However, in the QLSTM, quantum computing is only used to enhance the input data by transforming data via Variational Quantum Eigensolver.…”

Section: Quantum Neural Network For Sequential Inputmentioning

confidence: 99%

QNet: A Quantum-native Sequence Encoder Architecture

Day¹,

Chen²,

Sun³

2022

Preprint

View full text Add to dashboard Cite

This work investigates how current quantum computers can improve the performance of natural language processing tasks. To achieve this goal, we proposed QNet, a novel sequence encoder model entirely inferences on the quantum computer using a minimum number of qubits. QNet is inspired by Transformer, the state-of-the-art neural network model based on the attention mechanism to relate the tokens. While the attention mechanism requires time complexity of O(n 2 • d) to perform matrix multiplication operations, QNet has merely O(n + d) quantum circuit depth, where n and d represent the length of the sequence and the embedding size, respectively. To employ QNet on the NISQ devices, ResQNet, a quantum-classical hybrid model composed of several QNet blocks linked by residual connections, is introduced. We evaluate ResQNet on various natural language processing tasks, including text classification, rating score prediction, and named entity recognition. ResQNet exhibits a 6% to 818% performance gain on all these tasks over classical state-of-the-art models using the exact embedding dimensions. In summary, this work demonstrates the advantage of quantum computing in natural language processing tasks.

show abstract

Quantum Long Short-Term Memory

Cited by 55 publications

References 27 publications

Rapid training of quantum recurrent neural network

Rapid training of quantum recurrent neural network

IQGAN: Robust Quantum Generative Adversarial Network for Image Synthesis On NISQ Devices

QNet: A Quantum-native Sequence Encoder Architecture

Contact Info

Product

Resources

About