How IBM wants to democratize quantum computing via the cloud

The US group’s offer has brought together 40,000 users worldwide in just over a year. For particularly varied use cases.

Since its emergence in the 70s, quantum computing has fantasized more than one developer. Our computers reason in a binary way, linking the 0s and the 1s. Adding memory or giving them more and more powerful processors can not – despite Moore’s law – allow them to solve the most complex problems. With the quantum computer, we change dimension. As its name suggests, it uses the properties of quantum physics and especially the principle of “superposition” of the states of a particle. Unlike the bit, the qubit, the unit of measurement of quantum computing, can simultaneously represent a 0 and a 1. Dedicated to combinatorial analysis, quantum computation lends itself to specific cases of use in particular fields. of artificial intelligence or cryptography.

Until recently, quantum computing was restricted to R & D departments alone. In fact, it requires the adjustment of particularly high technical constraints to ensure qubit stability. Isolated from the outside world, the system must be protected from magnetic interference and cooled to temperatures close to absolute zero (-273.15 ° C). The slightest change in the environment can change a value and lead to a miscalculation.

A service launched in May 2016

Suffice to say that quantum computing is not within the reach of everyone. Or rather was. Once again, the cloud is changing the game. As it was able to do with artificial intelligence via APIs , it makes accessible to as many technologies hitherto reserved only to experts. In any case, this is the bet that IBM makes. Since May 2016, Big Blue has been offering users who want to experiment on a quantum machine accessible in cloud mode. While it is physically installed on the TJ Watson research center of the American IT giant in New York, anyone on the planet can enjoy its remote computing power.

Called IBM Q, this service has already attracted 40,000 customers in a hundred countries. 300,000 quantum experiments have been conducted worldwide via this infrastructure, and 15 scientific papers  have been published in the wake – beyond those proposed by IBM.

A free version

IBM started with a five-qubit processor to ramp up. IBM Q is based on two configurations today. Accessible for free, the first is based on a 16-bit processor. It is intended for developers and researchers who want to run algorithms, explore tutorials, perform simulations. A software development kit is available on GitHub . A good way to build an ecosystem, one of the objectives posted by IBM. The second configuration relies on a 17-bit processor, the most powerful processor created to date by the group. The hardware architecture has also been improved. This time it is a commercial offer whose pricing is defined on a case by case basis depending on the project.

And the computer giant does not stop there. In the next decade, he intends to reach 50 qubits and design a universal quantum computer, capable of performing all kinds of calculations. Well beyond the power of the most powerful supercomputers. A digital Grail that also covets its competitors.

In May 2013, Google launched the Quantum Artificial Intelligence Lab, a research laboratory focused exclusively on quantum computing. In late 2015, Google introduced D-Wave 2X, a new quantum computer developed by the manufacturer D-Wave and of which he is co-owner with NASA. Microsoft has been working on the problem since 2005 and the creation of Station Q , a dedicated R & D center.

Combination of quantum computing and AI

Technical Director of the Systems Hardware Division at IBM, Xavier Vasques welcomes the progress made. “This is another way of designing computers, for example we are using materials experts, but if we stay in R & D, we have taken a step in industrializing our processes and making this computing power available in the cloud. “, he welcomes. According to him, the use cases that are emerging are numerous and varied. “In health, quantum computing can help pharmaceutical research by studying how the molecules that make up a drug interact with one another.” The generalized sequencing of DNA via the

Other avenues mentioned include the optimization of supply chains in supply chain management and risk analysis to optimize financial investments.

Endowed with infinite computing power, quantum computing combined with artificial intelligence could open up new perspectives. If for the time being IBM Q is not associated with the platform of AI Watson of Big Blue, it should be it tomorrow. Finally, these advances will certainly help to secure the cloud. Specific to quantum computing, Grover’s and Shor’s AI algorithms already provide superior protection over current RSA-based encryption.