The output distributions of quantum circuits are an interesting class of probability distributions. They feature very prominently in the quest for quantum advantage via sampling experiments. Recently, they have also been of great interest in Quantum Machine Learning: they constitute a very versatile class of models for unsupervised machine learning that can be trained by updating the parameters of a parametrized quantum circuit. But how hard is it to learn these distributions from data? Can the training be done efficiently? In this talk, I will introduce you to a learning theory based perspective on these questions. This perspective allows us to deduce rigorous and quite strong answers while also bringing up many further questions.
Zoom meeting details
Zoom link: https://us06web.zoom.us/j/87850698108
If you encounter any problems with connecting to the Zoom meeting, please email firstname.lastname@example.org.