End-to-End Quantum Machine Learning Implemented with Controlled Quantum Dynamics

Abstract

Toward quantum machine learning deployed on imperfect near-term intermediate-scale quantum (NISQ) processors, the entire physical implementation of should include as less as possible hand-designed modules with only a few ad-hoc parameters to be determined. This work presents such a hardware-friendly end-to-end quantum machine learning scheme that can be implemented with imperfect near-term intermediate-scale quantum (NISQ) processors. The proposal transforms the machine learning task to the optimization of controlled quantum dynamics, in which the learning model is parameterized by experimentally tunable control variables. Our design also enables automated feature selection by encoding the raw input to quantum states through agent control variables. Comparing with the gate-based parameterized quantum circuits, the proposed end-to-end quantum learning model is easy to implement as there are only few ad-hoc parameters to be determined. Numerical simulations on the benchmarking MNIST dataset demonstrate that the model can achieve high performance using only 3-5 qubits without downsizing the dataset, which shows great potential for accomplishing large-scale real-world learning tasks on NISQ processors.arning models. The scheme is promising for efficiently performing large-scale real-world learning tasks using intermediate-scale quantum processors.

arXiv comment10 pages, 6 figures

References in corpus (27)

• Quantum Computing in the NISQ era and beyond
• Supplementary information for "Quantum supremacy using a programmable superconducting processor"
• Quantum Machine Learning
• Hardware-efficient Variational Quantum Eigensolver for Small Molecules and Quantum Magnets
• Supervised learning with quantum enhanced feature spaces
• Quantum support vector machine for big data classification
• Quantum principal component analysis
• Microwave photonics with superconducting quantum circuits
• Coherent Josephson qubit suitable for scalable quantum integrated circuits
• Training Schrödinger's cat: quantum optimal control
• Quantum generative adversarial learning
• Hierarchical quantum classifiers
• Towards Quantum Machine Learning with Tensor Networks
• Unsupervised Machine Learning on a Hybrid Quantum Computer
• Training of Quantum Circuits on a Hybrid Quantum Computer
• Quantum generative adversarial learning in a superconducting quantum circuit
• Learning Robust and High-Precision Quantum Controls
• Quantum Algorithms for Fixed Qubit Architectures
• Maximum Principle Based Algorithms for Deep Learning
• Quantum approximate optimization is computationally universal
• Quantum embeddings for machine learning
• Generative model benchmarks for superconducting qubits
• An Optimal Control Approach to Deep Learning and Applications to Discrete-Weight Neural Networks
• Quantum Kitchen Sinks: An algorithm for machine learning on near-term quantum computers
• Controlling qubit networks in polynomial time
• Robust Control Optimization for Quantum Approximate Optimization Algorithm
• Defining Quantum Neural Networks via Quantum Time Evolution

Cited by in corpus (10)

• Reinforcement Learning for Many-Body Ground-State Preparation Inspired by Counterdiabatic Driving
• Robust resource-efficient quantum variational ansatz through evolutionary algorithm
• Trainability Enhancement of Parameterized Quantum Circuits via Reduced-Domain Parameter Initialization
• Supervised learning for robust quantum control in composite-pulse systems
• Automatic Depth Optimization for Quantum Approximate Optimization Algorithm
• Evaluating the Performance of Some Local Optimizers for Variational Quantum Classifiers
• Noise-Robust End-to-End Quantum Control using Deep Autoregressive Policy Networks
• Quantum compiling with a variational instruction set for accurate and fast quantum computing
• Quantum Graph Convolutional Neural Networks
• A Quantum Convolutional Neural Network for Image Classification