By Chris Mansell


Title: Suppressing quantum errors by scaling a surface code logical qubit
Organization: Google
The ultimate goal of quantum error correction is to create a fault-tolerant quantum computer. For this to happen, the error correcting mechanisms, which themselves are subject to the occasional error, must cross a threshold where they correct more errors than they introduce.  The surface code is designed to make several noisy physical qubits behave as a single, less noisy, so-called logical qubit. In some impressive new research, superconducting qubits are put into groups of either 9 or 25 in order to test whether the whole is less error-prone than its parts. The researchers found that their device was close to the fault-tolerance threshold. (Technically, increasing the size of the logical qubit did decrease the logical error rate but this was just a finite-size effect rather than a true indication of the scaling.) They estimated that they would need to improve component performance by 20% to get below the threshold and then make further improvements to have practical scaling.

Title: Transport of multispecies ion crystals through a junction in an RF Paul trap
Organization: Quantinuum
Trapping two different elements, such as Barium and Ytterbium, in an ion trap quantum computer brings several benefits. One is sympathetic cooling, where one species of ion is used as a qubit and the other species is subject to laser cooling. Since the two are coupled together through their Coulomb repulsion, the “qubit ion” is cooled down without directly interacting with the decoherence-inducing cooling beams. Another advantage is that the measurement efficiency is extremely high. However, for the ion trap to be scalable, the ions need to be able to move between different regions of the trap. Now, for the first time, this movement has been demonstrated for a dual-species ion crystal. The pair of ions could successfully travel at 4 meters per second through a junction in the trap architecture.

Title: Optimal Purification of a Spin Ensemble by Quantum-Algorithmic Feedback
Organizations: University of Cambridge; University of Sheffield; Université Côte d’Azur
In this impressive experiment, a single electron confined in a semiconductor quantum dot caused approximately 100,000 nuclei to approach a target state that could be determined to within one nuclear spin flip. The thermal fluctuations were only a factor of 3 away from the fundamental quantum limit of single-spin fluctuations. 
This spectacular level of active control over the spin noise is important for several technological applications. Despite not being the focus of the paper, the record for the longest inhomogeneous dephasing time for any optically addressed quantum dot spin qubit was quadrupled. More importantly, the algorithmic approach that was demonstrated could lead to improved quantum memories and the ability to prepare quantum-correlated many-body states, such as a “Schrödinger kitten” state.

Title: A universal qudit quantum processor with trapped ions
Organizations: Universität Innsbruck; Österreichische Akademie der Wissenschaften; Alpine Quantum Technologies GmbH
Bits have a long history in classical information processing, from the punched cards that controlled looms in the 18th century to the dots and dashes of Morse code a hundred years later and eventually to modern, electronic computers. The practicalities that are currently shaping the development of quantum processors are somewhat different, to say the least. The vast majority of quantum systems have more than two possible quantum states, which means that qubits are not the only option for encoding data into the processor. If multiple energy levels are employed, the system is said to consist of qudits. The advantages of qudits over qubits are manifold but their use comes with technical challenges. Most implementations to date have been proof-of-principle demonstrations but in this latest article, trapped ions making use of up to seven energy levels each enact a universal gate set and perform as well as a comparable qubit system.


Title: Verifiable Quantum Advantage without Structure
Organizations: NTT Social Informatics Laboratories; Princeton University; NTT Research
Understanding this breakthrough requires a fair amount of background. Most readers will know that a quantum advantage is when a quantum computer answers a problem with a dramatically better runtime than a classical computer and that verifiability means it is easy to check that the answer is correct. The structure that is mentioned in the title of the paper refers to the distinction between periodic functions and random functions. The former are important to famous quantum algorithms such as Shor’s, while the latter are important in cryptography. The authors of the paper devised their own unusual problem to which several valid answers are possible. They considered a quantum processor that can give a superposition of inputs to a random oracle function and a normal computer that can only access the function classically. They showed that this setup produces a verifiable quantum advantage despite there not being any structure in the problem that the quantum computer could exploit.

Title: Modular Parity Quantum Approximate Optimization
Organizations: University of Innsbruck; Parity Quantum Computing GmbH; Institute for Quantum Optics and Quantum Information of the Austrian Academy of Sciences
The Quantum Approximate Optimization Algorithm (QAOA) is a hybrid quantum-classical protocol of considerable promise. For example, in the absence of errors, its performance is guaranteed to increase with the depth of the quantum circuit. Parity QAOA is an approach that allows constrained optimization problems to be solved on quantum devices where the qubits have limited connectivity. Previously, two different methods of implementing the constraints have been considered, each with its own pros and cons. The present work shows that these methods can be combined in an adjustable way so that for a given problem and a given noisy quantum processor, the most suitable circuit depth can be chosen. This has the additional effect that the procedure becomes very modular and scalable.

Title: Towards Quantum Advantage in Financial Market Risk using Quantum Gradient Algorithms
Organizations: Goldman, Sachs & Co.; IBM
Financial firms have to hedge their exposure to market risk by computing how sensitive the prices of their financial derivatives are to changes in model and market parameters. In many cases, there is no closed form for the price, ruling out an analytical approach. Instead, Monte Carlo simulations that require a lot of compute are used. Recent results have shown that quantum amplitude estimation leads to better scaling than classical Monte Carlo methods for very similar financial calculations. Inspired by this, the authors of this newest work investigate whether sensitivity analyses can be sped up by quantum algorithms for finding gradients. They discuss the required resources in terms of the clock rate for quantum logic gates and whether the algorithm can be implemented in parallel across multiple quantum processing units. They find a more promising path to quantum advantage than was found in the earlier related studies.

Title: Quantum operations with indefinite time direction
Organizations: The University of Hong Kong; University of Oxford; Perimeter Institute for Theoretical Physics
Quantum experiments often take the same form: prepare a quantum system, let time go by and then make some measurements. Curiously, our best understanding of the laws of nature is that they are time-symmetric but when we let time pass, it always passes in the direction that we call forwards. How would our information processing abilities be different if, hypothetically, we could choose the future state of a quantum system that somehow evolves backwards in time? Such head-scratchers are beyond most people’s patience levels, yet physicists can glean great insights from them. Just consider the huge impact that Maxwell’s demon had on our understanding of thermodynamics to see the importance of unusual thought experiments. The authors of the current work establish a mathematical framework for analysing scenarios where inputs and outputs are inverted. It extends and generalises prior work involving superpositions of processes happening in different orders.

July 28, 2022