-
Quantum Federated Learning with Entanglement Controlled Circuits and Superposition Coding
Authors:
Won Joon Yun,
Jae Pyoung Kim,
Hankyul Baek,
Soyi Jung,
Jihong Park,
Mehdi Bennis,
Joongheon Kim
Abstract:
While witnessing the noisy intermediate-scale quantum (NISQ) era and beyond, quantum federated learning (QFL) has recently become an emerging field of study. In QFL, each quantum computer or device locally trains its quantum neural network (QNN) with trainable gates, and communicates only these gate parameters over classical channels, without costly quantum communications. Towards enabling QFL und…
▽ More
While witnessing the noisy intermediate-scale quantum (NISQ) era and beyond, quantum federated learning (QFL) has recently become an emerging field of study. In QFL, each quantum computer or device locally trains its quantum neural network (QNN) with trainable gates, and communicates only these gate parameters over classical channels, without costly quantum communications. Towards enabling QFL under various channel conditions, in this article we develop a depth-controllable architecture of entangled slimmable quantum neural networks (eSQNNs), and propose an entangled slimmable QFL (eSQFL) that communicates the superposition-coded parameters of eS-QNNs. Compared to the existing depth-fixed QNNs, training the depth-controllable eSQNN architecture is more challenging due to high entanglement entropy and inter-depth interference, which are mitigated by introducing entanglement controlled universal (CU) gates and an inplace fidelity distillation (IPFD) regularizer penalizing inter-depth quantum state differences, respectively. Furthermore, we optimize the superposition coding power allocation by deriving and minimizing the convergence bound of eSQFL. In an image classification task, extensive simulations corroborate the effectiveness of eSQFL in terms of prediction accuracy, fidelity, and entropy compared to Vanilla QFL as well as under different channel conditions and various data distributions.
△ Less
Submitted 3 December, 2022;
originally announced December 2022.
-
Quantum Split Neural Network Learning using Cross-Channel Pooling
Authors:
Won Joon Yun,
Hankyul Baek,
Joongheon Kim
Abstract:
In recent years, the field of quantum science has attracted significant interest across various disciplines, including quantum machine learning, quantum communication, and quantum computing. Among these emerging areas, quantum federated learning (QFL) has gained particular attention due to the integration of quantum neural networks (QNNs) with traditional federated learning (FL) techniques. In thi…
▽ More
In recent years, the field of quantum science has attracted significant interest across various disciplines, including quantum machine learning, quantum communication, and quantum computing. Among these emerging areas, quantum federated learning (QFL) has gained particular attention due to the integration of quantum neural networks (QNNs) with traditional federated learning (FL) techniques. In this study, a novel approach entitled quantum split learning (QSL) is presented, which represents an advanced extension of classical split learning. Previous research in classical computing has demonstrated numerous advantages of split learning, such as accelerated convergence, reduced communication costs, and enhanced privacy protection. To maximize the potential of QSL, cross-channel pooling is introduced, a technique that capitalizes on the distinctive properties of quantum state tomography facilitated by QNNs. Through rigorous numerical analysis, evidence is provided that QSL not only achieves a 1.64\% higher top-1 accuracy compared to QFL but also demonstrates robust privacy preservation in the context of the MNIST classification task.
△ Less
Submitted 8 April, 2023; v1 submitted 11 November, 2022;
originally announced November 2022.
-
Projection Valued Measure-based Quantum Machine Learning for Multi-Class Classification
Authors:
Won Joon Yun,
Hankyul Baek,
Joongheon Kim
Abstract:
In recent years, quantum machine learning (QML) has been actively used for various tasks, e.g., classification, reinforcement learning, and adversarial learning. However, these QML studies are unable to carry out complex tasks due to scalability issues on input and output which is currently the biggest hurdle in QML. Therefore, the purpose of this paper is to overcome the problem of scalability. M…
▽ More
In recent years, quantum machine learning (QML) has been actively used for various tasks, e.g., classification, reinforcement learning, and adversarial learning. However, these QML studies are unable to carry out complex tasks due to scalability issues on input and output which is currently the biggest hurdle in QML. Therefore, the purpose of this paper is to overcome the problem of scalability. Motivated by this challenge, we focus on projection-valued measurements (PVM) which utilize the nature of probability amplitude in quantum statistical mechanics. By leveraging PVM, the output dimension is expanded from $q$, which is the number of qubits, to $2^q$. We propose a novel QML framework that utilizes PVM for multi-class classification. Our framework is proven to outperform the state-of-the-art (SOTA) methodologies with various datasets, assuming no more than 6 qubits are used. Furthermore, our PVM-based QML shows about $42.2\%$ better performance than the SOTA framework.
△ Less
Submitted 14 November, 2022; v1 submitted 29 October, 2022;
originally announced October 2022.
-
3D Scalable Quantum Convolutional Neural Networks for Point Cloud Data Processing in Classification Applications
Authors:
Hankyul Baek,
Won Joon Yun,
Joongheon Kim
Abstract:
With the beginning of the noisy intermediate-scale quantum (NISQ) era, a quantum neural network (QNN) has recently emerged as a solution for several specific problems that classical neural networks cannot solve. Moreover, a quantum convolutional neural network (QCNN) is the quantum-version of CNN because it can process high-dimensional vector inputs in contrast to QNN. However, due to the nature o…
▽ More
With the beginning of the noisy intermediate-scale quantum (NISQ) era, a quantum neural network (QNN) has recently emerged as a solution for several specific problems that classical neural networks cannot solve. Moreover, a quantum convolutional neural network (QCNN) is the quantum-version of CNN because it can process high-dimensional vector inputs in contrast to QNN. However, due to the nature of quantum computing, it is difficult to scale up the QCNN to extract a sufficient number of features due to barren plateaus. Motivated by this, a novel 3D scalable QCNN (sQCNN-3D) is proposed for point cloud data processing in classification applications. Furthermore, reverse fidelity training (RF-Train) is additionally considered on top of sQCNN-3D for diversifying features with a limited number of qubits using the fidelity of quantum computing. Our data-intensive performance evaluation verifies that the proposed algorithm achieves desired performance.
△ Less
Submitted 18 October, 2022;
originally announced October 2022.
-
Scalable Quantum Convolutional Neural Networks
Authors:
Hankyul Baek,
Won Joon Yun,
Joongheon Kim
Abstract:
With the beginning of the noisy intermediate-scale quantum (NISQ) era, quantum neural network (QNN) has recently emerged as a solution for the problems that classical neural networks cannot solve. Moreover, QCNN is attracting attention as the next generation of QNN because it can process high-dimensional vector input. However, due to the nature of quantum computing, it is difficult for the classic…
▽ More
With the beginning of the noisy intermediate-scale quantum (NISQ) era, quantum neural network (QNN) has recently emerged as a solution for the problems that classical neural networks cannot solve. Moreover, QCNN is attracting attention as the next generation of QNN because it can process high-dimensional vector input. However, due to the nature of quantum computing, it is difficult for the classical QCNN to extract a sufficient number of features. Motivated by this, we propose a new version of QCNN, named scalable quantum convolutional neural network (sQCNN). In addition, using the fidelity of QC, we propose an sQCNN training algorithm named reverse fidelity training (RF-Train) that maximizes the performance of sQCNN.
△ Less
Submitted 11 December, 2022; v1 submitted 25 September, 2022;
originally announced September 2022.
-
FV-Train: Quantum Convolutional Neural Network Training with a Finite Number of Qubits by Extracting Diverse Features
Authors:
Hankyul Baek,
Won Joon Yun,
Joongheon Kim
Abstract:
Quantum convolutional neural network (QCNN) has just become as an emerging research topic as we experience the noisy intermediate-scale quantum (NISQ) era and beyond. As convolutional filters in QCNN extract intrinsic feature using quantum-based ansatz, it should use only finite number of qubits to prevent barren plateaus, and it introduces the lack of the feature information. In this paper, we pr…
▽ More
Quantum convolutional neural network (QCNN) has just become as an emerging research topic as we experience the noisy intermediate-scale quantum (NISQ) era and beyond. As convolutional filters in QCNN extract intrinsic feature using quantum-based ansatz, it should use only finite number of qubits to prevent barren plateaus, and it introduces the lack of the feature information. In this paper, we propose a novel QCNN training algorithm to optimize feature extraction while using only a finite number of qubits, which is called fidelity-variation training (FV-Training).
△ Less
Submitted 18 September, 2022;
originally announced September 2022.
