By Dr. Chris Mansell

Shown below are summaries of a few interesting research papers related to quantum computing that have been published over the past month.

Hardware

Title: Spin Readout of a CMOS Quantum Dot by Gate Reflectometry and Spin-Dependent Tunneling 
Organizations: UCL, Quantum Motion Technologies, CEA Leti Institute, Hitachi Cambridge Laboratory and Université Grenoble Alpes 
For decades, CMOS (complementary metal oxide semiconductor) manufacturing has been the dominant way of fabricating integrated circuits. This group of researchers have demonstrated that the same highly scalable processes can be used to make qubits that are long-lived and have a small footprint. You can view the paper here: https://journals.aps.org/prxquantum/abstract/10.1103/PRXQuantum.2.010353

Title: Optimal operation points for ultrafast, highly coherent Ge hole spin-orbit qubits
Organizations: The University of New South Wales and the University of British Columbia
Germanium has a strong and tunable spin-orbit coupling, which enables qubit control via electric fields. Does this enhance undesired couplings to stray fields and lead to quick relaxation and dephasing? The authors demonstrate theoretically that this is emphatically not the case: Germanium actually exhibits an optimal operation point at which sensitivity to charge noise is minimised and the speed of electrical operation is maximised. This paper helps overturn the conventional wisdom that fast operation implies reduced lifetimes. You can view the paper here: https://www.nature.com/articles/s41534-021-00386-2

Title: Demonstration of the trapped-ion quantum CCD computer architecture
Organizations: Honeywell Quantum Solutions
In this work, various experimental achievements are incorporated into a single device which, because of its similarity to charge-coupled device (CCD) arrays, is called a QCCD quantum computer. While the road ahead still has some unmet challenges, this performant and modular approach certainly helps pave the way towards devices with thousands of qubits. You can view the paper here https://www.nature.com/articles/s41586-021-03318-4 or the arXiv version here https://arxiv.org/abs/2003.01293

Title: Categorizing Readout Error Correlations on Near Term Quantum Computers
Organizations: Lawrence Berkeley National Laboratory and the University of Georgia
Some error-correction schemes assume that the errors that occur in quantum computers are uncorrelated. This paper investigates the extent to which this true in two IBM devices. It finds that for both of them, the error correlations are small compared to the single-qubit readout errors. While the  correlations on the smaller device are long-ranged and do not decay with inter-qubit distance, for the larger, newer device, this is not the case. You can view the paper here: https://arxiv.org/abs/2104.04607

Title: Scalable Benchmarks for Gate-Based Quantum Computers
Organizations: QuSoft, University of Amsterdam; CQIF, University of Cambridge; and IQIM, California Institute of Technology
Businesses want to be able to compare different computers against each other and ensure that the one they purchase is able to fulfil its intended role. While various metrics exist for classical computers, they are few and far between for quantum computers. This paper details a suit of quantum benchmarks that can facilitate comparisons of both current and near future quantum platforms from an end-user perspective. You can view the paper here: https://arxiv.org/abs/2104.10698

Software

Title: Variational inference with a quantum computer
Organizations: Cambridge Quantum Computing and the University of Edinburgh
When new evidence comes to light, people readjust their confidence in their suspicions or hypotheses. This is known as inference and it can be performed rigorously by computers. The authors simulate near-term quantum Born machines and discover that they may be extremely well-suited to the inferential tasks that occur in many commercial industries. You can view the paper here: https://arxiv.org/abs/2103.06720

Title: Towards understanding the power of quantum kernels in the NISQ era
Organizations: Wuhan University and JD Explore Academy
Some theoretical results on the usefulness of quantum machine learning algorithms are established in perfectly noiseless settings. This paper considers the question of whether quantum versions of kernel methods will provide any advantages when run on NISQ devices. Even though the immediate answer is that they won’t, the authors describe a strategy to suppress estimation errors so that the performance from the ideal scenario can be maintained. You can view the paper here: https://arxiv.org/abs/2103.16774

Title: Equivalence of quantum barren plateaus to cost concentration and narrow gorges
Organizations: Los Alamos National Laboratory
Optimisers have to explore cost function landscapes in order to find the deepest valley (i.e. the global minimum). Could quantum landscapes be different from those encountered in classical optimisation? The authors discovered that certain mathematically possible landscapes are ruled out by quantum mechanics. With this knowledge, so-called “quantum-aware” optimisers can now navigate more efficiently. You can view the paper here: https://arxiv.org/abs/2104.05868

Title: Quantum agents in the Gym: a variational quantum algorithm for deep Q-learning
Organizations: Leiden University, Volkswagen and the University of Innsbruck
In classical computing, Q-learning is a popular and successful reinforcement learning algorithm. In this work, a quantum version of the Q-learning protocol is found to be competitive with a classical neural network in a simple benchmark environment. The possibility is discussed that in a certain special case, learning advantages over classical agents could be guaranteed. You can view the paper here: https://arxiv.org/abs/2103.15084

Title: Quantum versus classical generative modelling in finance
Organizations: The University of Edinburgh, Sorbonne Université and Rigetti Computing
This paper pits two protocols against each other. In the classical corner is a Boltzmann machine and in the quantum corner is a Born machine. The task is that of synthetic data generation. The Born machine comes out on top. Interestingly, its entangling capability is positively correlated with the amount by which it beats the Boltzmann machine. You can view the paper here: https://iopscience.iop.org/article/10.1088/2058-9565/abd3db

Title: Doubling the size of quantum simulators by entanglement forging
Organizations: IBM
Classical and quantum computers must work in tandem but understanding the best ways to do this is challenging. These researchers devise a procedure that, under certain conditions, allows an ordinary computer and n qubits to simulate a 2n-qubit quantum computation. They employ their procedure to compute the ground state energy of a water molecule in the most accurate quantum simulation of water to date. You can view the paper here: https://arxiv.org/abs/2104.10220

Title: The XZZX surface code
Organizations: University of Sydney and AWS
When a second-year physics undergraduate made a simple change to an error-correction code that had been thoroughly analysed for almost two decades, it doubled its ability to suppress errors. Future research will investigate whether such drastic improvements also occur when the same trick is applied to other codes that deal with biased errors. You can view the paper here: https://www.nature.com/articles/s41467-021-22274-1

April 29, 2021