Google has published free, open source software that will facilitate the creation of quantum programs. TensorFlow Quantum is an addition to the popular Google TensorFlow toolkit, which has helped give impetus to the development of machine learning since its launch in 2015.
TensorFlow Quantum will allow you to create quantum programs without diving into the details of the equipment on which they work. The user will be able to switch between a real quantum computer and a simulator. This means that you can first check how the program works in the simulation before running it on a full-scale quantum installation. Masood Mohseni, who leads the TensorFlow Quantum project, expects programmers to use the software to discover new algorithms that the company can use.
This is not the first set of tools for quantum machine learning. For example, the Xanadu startup in Toronto offers a similar platform called Pennylane. Google notes that they are ready to cooperate with such initiatives – according to them, this allows you to exchange ideas, which contributes to innovation. The Google team wants most quantum power researchers to use TensorFlow Quantum.
Previously, Google created the most powerful quantum computer that can perform calculations beyond the capabilities of other devices. At first she published a presentation about this, but later deleted information about “quantum superiority.” This means that the company’s quantum devices are capable of performing tasks that are not available even to the most powerful supercomputer.