Quantum Machine Learning

Training a model on IBM quantum computers with the Iris flower dataset

We have a new algorithmic approach for doing machine learning with quantum computers. We trained our qmodel for the ternary classification of the Iris flower dataset on IBM quantum computers. It reaches the accuracy level of classical ML.

PolyadicQML, our python library for definition, training and deployment of quantum models, is available on GitHub and ready for install on PyPI. Start your own QML project with a sample qmodel.

> pip install polyadicqml        

To get a technical grasp of the algorithm see our presentation at IBM Supercomputing virtual event. Otherwise a medium post is available for non-technical public. For more in deep details read the preprint of the research paper to be presented at QCE20.

Explore the training data:

Quantum Circuit
The quantum circuit corresponding to the qmodel. The θi are the parameters to learn, ωi are the input values.

A 15-min talk for the IBM Singapore Supercomputing virtual event held on June 30th, 2020.

The slides are available for download.