An exponentially-growing family of universal quantum circuits

Abstract

Quantum machine learning has become an area of growing interest but has certain theoretical and hardware-specific limitations. Notably, the problem of vanishing gradients, or barren plateaus, renders the training impossible for circuits with high qubit counts, imposing a limit on the number of qubits that data scientists can use for solving problems. Independently, angle-embedded supervised quantum neural networks were shown to produce truncated Fourier series with a degree directly dependent on two factors: the depth of the encoding and the number of parallel qubits the encoding applied to. The degree of the Fourier series limits the model expressivity. This work introduces two new architectures whose Fourier degrees grow exponentially: the sequential and parallel exponential quantum machine learning architectures. This is done by efficiently using the available Hilbert space when encoding, increasing the expressivity of the quantum encoding. Therefore, the exponential growth allows staying at the low-qubit limit to create highly expressive circuits avoiding barren plateaus. Practically, parallel exponential architecture was shown to outperform the existing linear architectures by reducing their final mean square error value by up to 44.7% in a one-dimensional test problem. Furthermore, the feasibility of this technique was also shown on a trapped ion quantum processing unit.

14 pages, 7 figures

References in corpus (13)

• Quantum algorithm for solving linear systems of equations
• Beyond Barren Plateaus: Quantum Variational Algorithms Are Swamped With Traps
• Is quantum advantage the right goal for quantum machine learning?
• Supervised quantum machine learning models are kernel methods
• Quantum machine learning for image classification
• Hybrid quantum neural network for drug response prediction
• Quantum Machine Learning: from physics to software engineering
• Quantum algorithms applied to satellite mission planning for Earth observation
• Generalization despite overfitting in quantum machine learning models
• Benchmarking simulated and physical quantum processing units using quantum and hybrid algorithms
• Exponential data encoding for quantum supervised learning
• Parallel Hybrid Networks: an interplay between quantum and classical neural networks
• Practical application-specific advantage through hybrid quantum computing

Cited by in corpus (7)

• Quantum machine learning for image classification
• Hybrid quantum ResNet for car classification and its hyperparameter optimization
• Hybrid quantum physics-informed neural networks for simulating computational fluid dynamics in complex shapes
• Hybrid quantum cycle generative adversarial network for small molecule generation
• Quantum algorithms for scientific computing
• Forecasting steam mass flow in power plants using the parallel hybrid network
• Hybrid quantum recurrent neural network for remaining useful life prediction of turbofan engines