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.


Title: Quantum Information Scrambling on a Superconducting Qutrit Processor
Organizations: Lawrence Berkeley National Laboratory; University of California, Berkeley; Perimeter Institute for Theoretical Physics
Scrambling is a ubiquitous natural phenomenon where local information spreads out through all available degrees of freedom. The superconducting quantum computer described in this paper has sufficient capabilities to explore whether higher dimensional systems based on qutrits, instead of qubits, exhibit different scrambling modalities. As well as further investigation of this question, it could be used to perform qutrit versions of many important experiments, possibly including ones that demonstrate quantum supremacy.

Title: Experimental Deep Reinforcement Learning for Error-Robust Gateset Design on a Superconducting Quantum Computer
Organizations: Q-CTRL
Quantum computers are experimental systems that have many unknown and transient Hamiltonian terms with nonlinear dependencies on the applied, and possibly distorted, control signals. They also experience undesired crosstalk between the qubits and time-varying environmental noise. This paper shows that Deep Reinforcement Learning can be used to make quantum logic gates less susceptible to all these issues. For example, on an IBM quantum computer, the techniques enabled a gate to maintain a high fidelity despite being applied three times faster.

Title: CMOS-based cryogenic control of silicon quantum circuits
Organizations: Delft University of Technology; Netherlands Organization for Applied Scientific Research; Intel; École Polytechnique Fédérale de Lausanne
A cryogenic and integrated electronic control system would allow for large scale solid-state quantum computers. However, some semiconductor electronics do not operate well at low temperatures. The possible exception is CMOS. The authors of this paper created a CMOS control chip operating at 3 Kelvin that coherently drove silicon quantum bits with state-of-the-art fidelity, implemented several benchmarking protocols and performed a two-qubit version of the Deutsch-Josza algorithm.

Title: Resilience of Quantum Random Access Memory to Generic Noise
Organizations: Yale University; The University of Chicago
QRAM is a general-purpose architecture for the implementation of quantum oracles and can be understood as a generalization of classical RAM. For contrived noise models, the so-called “bucket-brigade” QRAM architecture is highly resistant to errors. This paper finds that it also resistant to more realistic noise models, which paves the way for small-scale, near-term experimental demonstrations of QRAM.


Title: OpenQASM 3: A broader and deeper quantum assembly language
Organizations: IBM; AWS; University of Sussex
The language used to specify quantum circuits in Qiskit is called OpenQASM. This paper describes some upgrades allow users to code a broad family of dynamic circuits, where there is greater interactivity between classical procedures, such as those that optimise or update various parameters, and the quantum circuits themselves. This may help to significantly improve the runtimes of algorithms like iterative phase estimation, repeat until success and chemical simulations.

Title: A Grand Unification of Quantum Algorithms
Organizations: Massachusetts Institute of Technology
In 2019, the quantum algorithms for search, factoring and simulation that had previously seemed quite disparate were seen through a unifying theoretical lens. It was based on a technique familiar to students of linear algebra called singular value decomposition. In this paper, the unification of quantum algorithms is explained and developed. These results could spur on a new age of progress for quantum algorithm design.

Title: Training Quantum Embedding Kernels on Near-Term Quantum Computers
Organizations: Freie Universität Berlin; University of Cologne; University of Copenhagen
Quantum classifiers have shown promising levels of robustness in the face of moderate circuit fidelity. This paper finds that even very simple error mitigation strategies be useful. This was tested across different circuit depths, qubit numbers and datasets. Open questions for further investigation include the effects of the barren plateaus when training kernel-based classifiers.

Title: Test of Quantumness with Small-Depth Quantum Circuits
Organizations: National Institute of Informatics; Nagoya University
The “learning with errors” task is to solve a system of “noisy” equations. Related tasks can be used as a test whether the inner workings of a device are harnessing quantum mechanical phenomena. This paper looks at the complexity of these tasks and finds that they can be solved by combining constant depth quantum circuits with logarithmically deep classical circuits. The result makes applications, such as producing certifiable randomness, seem feasible in the near term.

Title: Pauli error estimation via Population Recovery
Organizations: AWS; California Institute of Technology; Carnegie Mellon University
Researcher Steven T. Flammia once joked: all noiseless quantum computers are alike; every noisy quantum computer is noisy in its own way. The most important parameters for characterising a quantum device are the Pauli error error rates. He and coauthor Ryan O’Donnell devised an algorithm to efficiently estimate these rates. It is optimal up to the logarithmic factors and due to its simplicity, it can immediately be employed on present day devices.

Title: Quantum Advantage in Simulating Stochastic Processes
Organizations: Jagiellonian University; University of Gdańsk; The Barcelona Institute of Science and Technology; Delft University of Technology
Stochastic processes underlie a vast range of natural and social phenomena but simulating such systems often requires the use of a prohibitive amount of data. Quantum technologies have shown the potential to dramatically reduce the amount of working memory required to simulate stochastic processes. The authors found three scenarios where quantum advantages in memory or time arise. The potential applications seem as though they will be in the distant future rather than the near term but more research is needed.

Title: Quantum Advantage with Shallow Circuits under Arbitrary Corruption
Organizations: The University of Tokyo; Nagoya University
Most proofs of quantum speed ups rely on plausible assumptions from different fields of computer science. There is, however, a proof with no such assumptions that shallow quantum circuits can have a quantum advantage even if they are subject to local stochastic noise. This paper finds that the advantage persists under more severe noise. While there is no immediate practical application of this research, it advances the very important research topics of quantum advantages and error resilience.

Title: Quantum Monte-Carlo Integration: The Full Advantage in Minimal Circuit Depth
Organizations: Cambridge Quantum Computing; University of Cambridge
Monte Carlo methods are extremely useful. A key example is their use in the financial industry for option pricing. Due to the limited depth of NISQ circuits, it was thought that the speed up provided by a quantum version of such an algorithm might require a fault tolerant quantum computer. This latest work, as its title suggests, shows how the quantum circuit can have a small depth and still retain the full quantum advantage.