Mitigating Noise-Induced Gradient Vanishing in Variational Quantum Algorithm Training

Abstract

Variational quantum algorithms are expected to demonstrate the advantage of quantum computing on near-term noisy quantum computers. However, training such variational quantum algorithms suffers from gradient vanishing as the size of the algorithm increases. Previous work cannot handle the gradient vanishing induced by the inevitable noise effects on realistic quantum hardware. In this paper, we propose a novel training scheme to mitigate such noise-induced gradient vanishing. We first introduce a new cost function of which the gradients are significantly augmented by employing traceless observables in truncated subspace. We then prove that the same minimum can be reached by optimizing the original cost function with the gradients from the new cost function. Experiments show that our new training scheme is highly effective for major variational quantum algorithms of various tasks.

References in corpus (10)

• Supplementary information for "Quantum supremacy using a programmable superconducting processor"
• A Quantum Approximate Optimization Algorithm
• Noise-Induced Barren Plateaus in Variational Quantum Algorithms
• Absence of Barren Plateaus in Quantum Convolutional Neural Networks
• Layerwise learning for quantum neural networks
• Trainability of Dissipative Perceptron-Based Quantum Neural Networks
• Quantum computing and the entanglement frontier
• Large gradients via correlation in random parameterized quantum circuits
• Learning to learn with quantum neural networks via classical neural networks
• Toward Trainability of Quantum Neural Networks

Cited by in corpus (1)

• The Variational Quantum Eigensolver: a review of methods and best practices