THE NEED FOR QUANTUM SOFTWARE
Recent years have witnessed spectacular breakthroughs in the development of quantum computers.
In 2014, the UC Santa Barbara group led by John Martinis (now part of Google’s Quantum Artificial
Intelligence Lab
) created a high-performance 9-qubit platform. Subsequently they programmed it to
do the first-ever digital quantum simulation of a theoretical physics problem. Several groups are currently close to realizing quantum computers on 50 qubits.

The chart below shows the projected growth in the number of qubits in a quantum computer as projected at the start of QuSoft (though current developments are even faster than projected). It is based on the UCSB/Google developments sketched above, as well as progress on few-qubit platforms made at other labs using a variety of approaches.

qubitgrowth

Breaking boundaries

At the start of QuSoft we were at the 10-20 qubit scale, but soon researchers will cross the boundary of 50 qubits. At this point, ideas for quantum algorithms can be tested and the results can still be compared to classical programs.  This will allow quantum computations that would be practically impossible on a classical computer.

The exponential speed-up that a quantum machine can offer for some computational problems arises from its fundamentally different nature compared to classical computers. This in turn requires a fundamentally different architecture and fundamentally different programming. In terms of hardware, the birth moment of quantum computation is right around the corner, yet quantum algorithms and protocols are still few and far between.

Software: the key to success

Quantum software is, in fact, the key to success. Many groups are working on the hardware, but the questions of what a realistically-sized quantum computer can achieve, how do to this, and how to verify the results form a wide-open field of science. QuSoft is the clearest statement of this realization. At QuSoft, the urgent need for quantum software meets a unique combination of Amsterdam talent that spans computer science, mathematics and physics.

Which problems can be uniquely solved using a quantum algorithm based on small-qubit ensembles? Here theoretical physics is the right partner. How can qubits be used to make uncrackable codes that will remain secure in the quantum computer age? Here mathematicians and logicians are the people to look to. How should quantum software be tested, and how can we detect and remove bugs when classical machines can’t check the results? Here computer scientists bring the right expertise.

The software to be run actually defines the right way to build and connect the hardware. As a result, the architecture of the quantum computer is also central to QuSoft. This will be addressed by combining insights from computer science and physics.