Towards Efficient On-Chip Training of Quantum Neural Networks

29 Sep 2021  ·  Hanrui Wang, Zirui Li, Jiaqi Gu, Yongshan Ding, David Z. Pan, Song Han ·

Quantum Neural Network (QNN) is drawing increasingly more research interest thanks to its potential to achieve quantum advantage on near-term Noisy Intermediate Scale Quantum (NISQ) hardware. In order to achieve scalable QNN learning, the training process needs to be offloaded to real quantum machines instead of using exponential-cost classical simulators. One common approach to obtain QNN gradients is parameter shift whose cost scales linearly with the number of qubits. This work presents the first experimental demonstration of practical on-chip QNN training with parameter shift. Nevertheless, we find that due to the significant quantum errors (noises) on real machines, gradients obtained from naive parameter shift have low fidelity and thus degrade 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 on 5 classification tasks with 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% \qnn accuracy improvements over no pruning. Overall, we successfully obtain comparable accuracy with noise-free simulation but have much better training scalability. We also open-source our PyTorch library for on-chip \qnn training with parameters shift and easy deployment at this link: https://anonymous.4open.science/r/iclr-on-chip-qnn-572E .

PDF Abstract
No code implementations yet. Submit your code now

Datasets


Results from the Paper


  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.

Methods


No methods listed for this paper. Add relevant methods here