QOC: quantum on-chip training with parameter shift and gradient pruning

2022-02-26

Design Automation Conference

AI Summary

Training parameterized quantum circuits (PQCs) on Noisy Intermediate-Scale Quantum (NISQ) hardware is hindered by severe hardware noise, distorted gradients, and the infeasibility of classical simulation at scale. Method: This work introduces the first on-device PQC training system tailored for real quantum hardware. It enables end-to-end on-chip training—the first such demonstration on physical devices—and proposes a probabilistic gradient pruning mechanism that dynamically discards gradient components with high relative error based on magnitude, thereby enhancing noise robustness. Leveraging the parameter-shift rule and the TorchQuantum framework, the system efficiently deploys quantum neural network (QNN) training across five real quantum processors. Results: On binary and 4-class image classification tasks, it achieves >90% and >60% test accuracy, respectively; gradient pruning yields up to 7% absolute accuracy gain; training fidelity approaches ideal simulation performance, and the system exhibits linear scalability.

Abstract

Parameterized Quantum Circuits (PQC) are drawing increasing research interest thanks to its potential to achieve quantum advantages on near-term Noisy Intermediate Scale Quantum (NISQ) hardware. In order to achieve scalable PQC learning, the training process needs to be offloaded to real quantum machines instead of using exponential-cost classical simulators. One common approach to obtain PQC gradients is parameter shift whose cost scales linearly with the number of qubits. We present QOC, the first experimental demonstration of practical on-chip PQC training with parameter shift. Nevertheless, we find that due to the significant quantum errors (noises) on real machines, gradients obtained from naïve parameter shift have low fidelity and thus degrading the training accuracy. To this end, we further propose probabilistic gradient pruning to firstly identify gradients with potentially large errors and then remove them. Specifically, small gradients have larger relative errors than large ones, thus having a higher probability to be pruned. We perform extensive experiments with the Quantum Neural Network (QNN) benchmarks on 5 classification tasks using 5 real quantum machines. The results demonstrate that our on-chip training achieves over 90% and 60% accuracy for 2-class and 4-class image classification tasks. The probabilistic gradient pruning brings up to 7% PQC accuracy improvements over no pruning. Overall, we successfully obtain similar on-chip training accuracy compared with noise-free simulation but have much better training scalability. The QOC code is available in the TorchQuantum library.

Problem

• Parameterized Quantum Circuits
• Training Accuracy
• Machine Errors

Innovation

• QOC Method
• Parameterized Quantum Circuits (PQC)
• Probability Gradient Pruning