Tutorial 4 – A trained generative model with qubits
In this forth chapter, we will develop a trained generative model , optimize the parameters of a quantum circuit and find out which role entanglement plays for learning complicated data distributions.
In this series of tutorials, we provide a basic introduction into some of the fundamental concepts behind unsupervised generative modelling with quantum hardware. Unsupervised machine learning algorithms have learned breathtaking tasks like the creation of photorealistic images from simple text inputs. Because of this great power, researchers started to translate those concepts to quantum hardware. In this series of tutorials on quantum technologies, we provide a basic introduction into some of the fundamental concepts behind generative modelling with quantum circuits.
Requirements
qiskit
.If you would like to have further information, please contact us.
Tutorial 1 – A baby example of generative modelling
In this introductory chapter, we will discuss the basic notions of generative modelling that work as a guideline for quantum algorithms. This is especially meant for a proper introduction of notations in the next few steps. Get started for free.
Tutorial 2 – Training generative models
In this second chapter, we will discuss how we train generative models to emulate the probabilistic distributions. This sets the stage for the quantum formulation in the next chapter.
Tutorial 3 – An untrained generative model with qubits
In this third chapter, we prepare a generative model with a few qubits. We introduce quantum circuits with a sufficiently high degree of freedom to generate interesting samples and show how continuous datasets are adapted to quantum circuits.
Tutorial 4 – A trained generative model with qubits
In this forth chapter, we will develop a trained generative model , optimize the parameters of a quantum circuit and find out which role entanglement plays for learning complicated data distributions.
About the instructor
Manuel Rudolph is currently a PhD student in Physics at EPFL with Prof. Zoë Holmes. Before that, he was a Quantum Application Scientist in Zapata Computing’s Quantum AI team, where he worked with Alejandro Perdomo-Ortiz on near-term QML algorithms.
Find him on Twitter, LinkedIn, or Google Scholar.