Verification of quantum technologies, applications & systems

Primary researchers involved VERIqTAS is a European consortium with the following partners: Remigiusz Augusiak (Center for Theoretical Physics, Polish Academy of Sciences), Antonio Acn (Institute of Photonic Sciences), Omar Fawzi (Centre de Recherche Inria Grenoble-Rhône-Alpes), Laura Mancinska (University of Copenhagen), Stefano Pironio (Université libre de Bruxelles) and Miguel Navascués (Institute of Quantum Optics and Quantum Information Vienna, Austrian Academy of Sciences). Quantum technologies have already been around for a while. Think of quantum key distribution: in 1984, Bennett and Brassard conceived a protocol that allowed the secret transmission of information between two separate locations. The security of their scheme, which required the two locations to send laser pulses to one another, was guaranteed by the laws of quantum mechanics. Many experiments followed, and in 2001, ID quantique, the oldest quantum cryptography provider, was founded. Nowadays there exists a quantum key distribution network connecting Beijing and Shanghai, and there are plans to build a similar one in the European Union. In parallel, in the last few years we have seen a proliferation of Noisy Intermediate-Scale Quantum (NISQ) devices. Such are complex quantum systems over which we have an extraordinary level of control, but which nonetheless are not sophisticated enough to make a quantum computer. Since conventional computers cannot predict the behavior NISQ devices, there is hope that they can be used to solve demanding computational tasks, such as medicine design. In fact, there already exist theoretical proposals to use NISQ devices as an aid in financial investment. On this beautiful blue sky there are, however, a few clouds: commercial QKD systems have been hacked repeatedly, and NISQ devices are known to sometimes output the wrong answers. In all cases, there was a hardware problem involved. Namely, one of the parts of the overall system, or their joint interaction, was not working as expected. The goal of project VERIqTAS is to develop protocols and methods to verify whether complex quantum systems, such as networks or NISQ devices, operate correctly. In the case of communication networks, this is all the more challenging, because we cannot even trust the experimental devices that we use to test the network: they could have been tampered by a spy! One would think impossible to verify that a system works without making assumptions on how we access the system. However, this is only true in classical physics: as it turns out, there exist quantum cryptographic schemes that do not rely on knowing how the experimental devices work. Such schemes are still very impractical, so we aim to improve them in project VERIqTAS. Verifying the correct functioning of NISQ devices is not much easier, because normal computers cannot simulate them. This means that whatever protocol we might think of cannot include a full model of the system that we wish to study. Project VERIqTAS will discern what features of complex quantum systems we can predict and use them to verify that the device is working.

In the last few years, we have witnessed a wave of novel quantum technologies, such as quantum random number generators, or the first prototypes of quantum computers. The use of these new devices for society relies on the correct functioning of their inner mechanisms. In this context, project Veriqtas sought to develop general techniques to certify that each quantum component in a complex device is doing what it is supposed to do. In this regard, the Austrian node started the project studying a minimal physical scenario involving just a preparation device and a measurement device. A preparation device is an apparatus that sets a quantum system, say, a photon, in a given configuration or state, e.g., horizontally polarized. A measurement device is an apparatus that, when fed a quantum system, returns an outcome that indicates some physical property of the system. For instance, the outcome of a measurement device could be "vertical", if it is measuring the polarization of the photon it interacts with. We considered the problem of certifying that a preparation device and a measurement device operate according to their specifications. Remarkably, we found that it is possible to certify that both devices work as expected without involving a third one, provided that we know the maximum number of bits of information that the quantum system can carry. Since this information is available in many experimental setups, one can use our methods to certify preparation and measurement devices under very mild assumptions. We next moved to the study of many-body quantum systems: quantum systems made of several independent parts. Such systems are difficult to analyze, because the number of parameters one needs to describe their state grows very quickly with the number of subsystems. We considered experiments where each part of the many-body system is measured individually: the outcome of all such measurements is combined to certify some non-trivial property of the system. In this framework, we found simple protocols to certify that the many-body system is close to zero temperature, or to predict how the system properties will change from one moment to the next. We even found a (simple) way to certify both the device preparing the many-body system and the individual measurement devices. Finally, we tackled the problem of certifying the security of quantum cryptographic protocols. In this regard, we obtained a negative result: we noted that many such protocols rely on the assumption that the quantum systems used to carry information can be fully reset; when one drops that assumption, though, the protocols cease to be secure. This hack affects many recently proposed protocols to generate certifiably random numbers.