This research presents a Dual Quantum-Traditional Auto encoder network designed for near-term quantum processors. This Dual AE consists of a variational Auto encoder and decoder network which are processed and trained using a minimal computer. In the encoding stage quantum circuits and angle based encoding method is utilized for implementation. Further the decoder model is implemented using differential two stage quantum encoding circuits. A compatible approach is developed for quantum networks thereby providing control over the accuracy balance and complexity of the circuit mechanism. By using this compatible approach scalability of the model can be improved and more qubits can be concerned in dataset dimensions. Models’ loss function is obtained for the implementation for quantum circuits in order to eliminating the need for more auxiliary bits. IBM Qis Kit development kit is utilized for simulating the process. However, a traditional computer is used for training and testing the variational auto encoder model by using Ada grad optimization algorithm. A dual traditional-quantum method is implemented using a four-stage network with varying qubit system. Proposed method is tested and the obtained results shows that the model achieve optimal performance which includes 68 gates and various trainable parameters. Minimal divergence is obtained between real data when compared with generated ones.
This research presents a Dual Quantum-Traditional Auto encoder network designed for near-term quantum processors. This Dual AE consists of a variational Auto encoder and decoder network which are processed and trained using a minimal computer. In the encoding stage quantum circuits and angle based encoding method is utilized for implementation. Further the decoder model is implemented using differential two stage quantum encoding circuits. A compatible approach is developed for quantum networks thereby providing control over the accuracy balance and complexity of the circuit mechanism. By using this compatible approach scalability of the model can be improved and more qubits can be concerned in dataset dimensions. Models’ loss function is obtained for the implementation for quantum circuits in order to eliminating the need for more auxiliary bits. IBM Qis Kit development kit is utilized for simulating the process. However, a traditional computer is used for training and testing the variational auto encoder model by using Ada grad optimization algorithm. A dual traditional-quantum method is implemented using a four-stage network with varying qubit system. Proposed method is tested and the obtained results shows that the model achieve optimal performance which includes 68 gates and various trainable parameters. Minimal divergence is obtained between real data when compared with generated ones.
In this article, we present a hybrid quantum-classical generative adversarial network (GAN) for near-term quantum processors. The hybrid GAN comprises a variational generator and a discriminator quantum neural network, which are trained using a classic computer. The generator network is realized using angle-encoding and variational quantum circuits. The discriminator network is realized using multi-stage trainable encoding quantum circuits. A modular design approach is proposed for quantum neural networks which allows control on their depth to compromise accuracy and circuit complexity. Moreover, this modular approach makes the quantum neural networks amenable to scaling up the dimension of dataset and the number of qubits. Gradients of the loss functions are derived using the same quantum circuits used for the implementation of quantum neural networks to prevent the need for extra quantum circuits or auxiliary qubits. The quantum simulations are performed using the IBM Qiskit open-source software development kit (SDK), while the training of the hybrid quantum-classical GAN is conducted using the mini-batch stochastic gradient descent (SGD) optimization on a classic computer. The hybrid quantum-classical GAN is realized using a two-qubit system with different discriminator network structures. The best performance is achieved using a five-stage discriminator network comprising 63 quantum gates and 31 trainable parameters, with the Kullback-Leibler (KL) and Jensen-Shannon (JS) divergence scores of 0.39 and 0.52, respectively, for the similarity between the real and generated data distributions.
In the context of current Noisy Intermediate-Scale Quantum (NISQ) computers, the application of quantum circuits is hampered by various sources of noise, resulting in errors between the actual output and the ideal output. To mitigate these errors, this study investigates a method for predicting the expected output of quantum circuits. In order to enhance the precision of the experiments, a fixed-scale quantum circuit dataset is randomly collected, and one-hot encoding is applied to represent the data. To eliminate redundant noise in the dataset, a Quantum auto-encoder (QAE) reconstruction encoding model is designed to separate and reconstruct the descriptions of quantum circuits in the dataset. The operation process of a quantum circuit is a sequence of quantum gate operations, and gate operations are gradually realized to change the state of qubits. We regard the process of quantum circuit implementation as a process of evolution with time, so it has the nature of timing. We build an additive fusion prediction model based on temporal neural network, the model can be trained to learn the relationship between sequences to predict the output of quantum circuit, and a loss function based on maximum likelihood estimation is employed to optimize the model parameters. Additionally, a dual approach is verified to optimize the computational indicators in the ensemble model. To validate the effectiveness of the proposed method, experiments are conducted on quantum circuits of varying scales, and the results are compared with the quantum system framework provided by IBM Quantum Experience. The experimental results show that the time series neural network integration can better highlight timing information and noise information in quantum circuits, the proposed method outperforms the IBM Quantum Experience framework in terms of circuit depth and network depth for quantum circuits of different scales, exhibiting robustness and stability.
A longstanding computational challenge is the accurate simulation of many-body particle systems. Especially for deriving key characteristics of high-impact but complex systems such as battery materials and high entropy alloys (HEA). While simple models allow for simulations of the required scale, these methods often fail to capture the complex dynamics that determine the characteristics. A long-theorized approach is to use quantum computers for this purpose, which allows for a more efficient encoding of quantum mechanical systems. In recent years, the field of quantum computing has become significantly more mature. Furthermore, the rise in integration of machine learning with quantum computing further pushes to a near-term advantage. In this work we aim to improve the well-established quantum computing method for calculating the inter-atomic potential, the variational quantum eigensolver, by presenting an auto-encoded VQE with neural-network predictions: NN-AE-VQE. We apply a quantum autoencoder for a compressed quantum state representation of the atomic system, to which a naive circuit ansatz is applied. This reduces the number of circuit parameters to optimize, while still minimal reduction in accuracy. Additionally, we train a classical neural network to predict the circuit parameters to avoid computationally expensive parameter optimization. We demonstrate these methods on a H2 molecule, achieving chemical accuracy. We believe this method shows promise of efficiently capturing highly accurate systems while omitting current bottlenecks of variational quantum algorithms. Finally, we explore options for exploiting the algorithm structure and further algorithm improvements.
Embedding of n-dimensional data in quantum space is an essential component in quantum computing realm. This research area is quite challenging with limited computational resources and complex quantum architecture. To overcome these problems, quantum approaches have been integrated with classical neural network. This hybrid model encodes the features using both angle encoding and amplitude encoding over quantum hilbert space, which is then processed through a Variational Quantum Circuit (VQC). These embedded features are fed to the classical neural network that enhances the performance of the traditional methods. A synthetic barcode and stripes (BAS) dataset was created for the verification of the proposed model. The results show that classical optimization methods, specifically gradient descent with momentum, achieve an accuracy of 82.38% for 4×4 data which is a significant improvement over existing researches that are limited to lower-dimensional data embedding.
This work is devoted to solving the problem of processing numerical sequences using variational quantum algorithms implemented on a noisy intermediate-scale quantum computer. The quantum computing platform is an integrated circuit based on superconducting artificial atoms. The architecture of the quantum recurrent neural network is constructed using single- and two-qubit operations. At the simulation stage, we investigated the trainability of the model as a function of the number of qubits, as well as the encoded into the qubit states data amount and the encoding method. A comparison with classical architectures showed that on current quantum processors it is possible to achieve a prediction quality on the chosen machine-learning task that is comparable to that of models implemented on classical processors.
We propose a quantum implicit neural representation (QINR)-based autoencoder (AE) and variational autoencoder (VAE) for image reconstruction and generation tasks. Our purpose is to demonstrate that the QINR in VAEs and AEs can transform information from the latent space into highly rich, periodic, and high-frequency features. Additionally, we aim to show that the QINR-VAE can be more stable than various quantum generative adversarial network (QGAN) models in image generation because it can address the low diversity problem. Our quantum-classical hybrid models consist of a classical convolutional neural network (CNN) encoder and a quantum-based QINR decoder. We train the QINR-AE/VAE with binary cross-entropy with logits (BCEWithLogits) as the reconstruction loss. For the QINR-VAE, we additionally employ Kullback-Leibler divergence for latent regularization with beta/capacity scheduling to prevent posterior collapse. We introduce learnable angle-scaling in data reuploading to address optimization challenges. We test our models on the MNIST, E-MNIST, and Fashion MNIST datasets to reconstruct and generate images. Our results demonstrate that the QINR structure in VAE can produce a wider variety of images with a small amount of data than various generative models that have been studied. We observe that the generated/reconstructed images from the QINR-VAE/AE are clear with sharp boundaries and details. Overall, we find that the addition of QINR-based quantum layers into the AE/VAE frameworks shows improved performance of reconstruction/generation under the constrained experimental setting relative to the specific baselines.