(TechnologyReview) Google’s TensorFlow is one of a number of tools that make machine learning more accessible, by simplifying deep neural networks and providing reusable code so that new machine-learning apps don’t have to be written from scratch. TensorFlow Quantum is set to do the same for quantum machine learning.
TensorFlow Quantum will let you write quantum apps without getting bogged down in the details of the hardware they are running on. A switch lets you flick between an actual quantum computer and a simulation of one on a classical machine. This means that you can debug your quantum app in a simulation before trying to run it on a full-blown quantum setup.
This is not the first toolkit for quantum machine learning. For example, quantum computing startup Xanadu in Toronto offers a similar platform called Pennylane. But it is still a big deal that Google is doing i, says Xanadu researcher Nathan Killoran. He notes that developers build communities around big-name tools like TensorFlow, sharing code and ideas, which fosters innovation. Machine-learning tech is in better shape today because of it.