Federated Quantum Machine Learning

Distributed training across several quantum computers could significantly improve the time and if we share learned model, not data, it potentially data privacy as would happen where is located. One of potential schemes to achieve this property federated learning (FL), which consists clients or local nodes on their own a central node aggregate models collected from those nodes. However, best our knowledge, no work has been done in machine (QML) federation setting yet. In work, present hybrid quantum-classical although framework be generalized pure model. Specifically, consider neural network (QNN) coupled with classical pre-trained convolutional Our distributed scheme demonstrated almost same level trained model accuracies yet faster training. It demonstrates promising future research direction for scaling aspects.