15. February 2018

Quantum entanglement is a key feature of a quantum computer. Yet, how can we verify that a quantum computer indeed incorporates a large-scale entanglement? Using conventional methods is hard since they require a large number of repeated measurements. Aleksandra Dimić from the University of Belgrade and Borivoje Dakić from the Austrian Academy of Sciences and the University of Vienna have developed a novel method where in many cases even a single experimental run suffices to prove the presence of entanglement. Their surprising results will be published in the online open access journal npj Quantum Information of the Nature Publishing group.

The ultimate goal of quantum information science is to develop a quantum computer, a fully-fledged controllable device which makes use of the quantum states of subatomic particles to store information. As with all quantum technologies, quantum computing is based on a peculiar feature of quantum mechanics, quantum entanglement. The basic units of quantum information, the qubits, need to correlate in this particular way in order for the quantum computer to achieve its full potential.

One of the main challenges is to make sure that a fully functional quantum computer is working as anticipated. In particular, scientists need to show that the large number of qubits are reliably entangled. Conventional methods require a large number of repeated measurements on the qubits for reliable verification. The more often a measurement run is repeated the more certain one can be about the presence of entanglement. Therefore, if one wants to benchmark entanglement in large quantum systems it will require a lot of resources and time, which is practically difficult or simply impossible. The main question arises: can we prove entanglement with only a low number of measurement trials?

Now researchers from the University of Belgrade, the University of Vienna and the Austrian Academy of Sciences have developed a novel verification method which requires significantly fewer resources and, in many cases, even only a single measurement run to prove large-scale entanglement with a high confidence. For Aleksandra Dimić from the University of Belgrade, the best way to understand this phenomenon is to use the following analogy: "Let us consider a machine which simultaneously tosses, say, ten coins. We manufactured the machine such that it should produce correlated coins. We now want to validate whether the machine produces the anticipated result. Imagine a single trial revealing all coins landing on tails. This is a clear signature of correlations, as ten independent coins have 0.01% chance to land on the same side simultaneously. From such an event, we certify the presence of correlations with more than 99.9% confidence. This situation is very similar to quantum correlations captured by entanglement." Borivoje Dakić says: "In contrast to classical coins, qubits can be measured in many, many different ways. The measurement result is still a sequence of zeros and ones, but its structure heavily depends on how we choose to measure individual qubits", he continues. "We realized that, if we pick these measurements in a peculiar way, entanglement will leave unique fingerprints in the measured pattern", he concludes.

The developed method promises a dramatic reduction in time and resources needed for reliable benchmark of future quantum devices.**Publication in npj Quantum Information:**

A.Dimić and B.Dakić, "Single-copy enntaglement detection", npj Quantum Information, 2018.

DOI: 10.1038/s41534-017-0055-x

http://www.nature.com/articles/s41534-017-0055-x