By Dr. Chris Mansell

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


Title: Quantum proton entanglement on a nanocrystalline silicon surface
Organizations: Nagoya City University; Chuo University; Japan Atomic Energy Agency; Shizuoka University; High Energy Accelerator Research Organization (KEK)
A major advantage of silicon qubits is that they could leverage decades of technology development in the semiconductor industry. In this paper, instead of investigating silicon’s electrons, the authors take an unconventional leap and search the quantum states of protons. They found an entangled state with a lifetime of more than 1 millisecond. The question is now whether they can integrate this into more mainstream plans for silicon quantum computers.

Title: Observation of Time-Crystalline Eigenstate Order on a Quantum Processor
Organizations: Google Quantum AI and collaborators
Ideas that are ahead of their time are sometimes ignored. This was not the case when in 2012 Frank Wilczek proposed time crystals, partly because he was already a Nobel laureate. Time crystals are states of matter that spontaneously and persistently display structure in time. They were first approximately realised a few years ago but now a team from Google have examined one in four different ways. They found that the main reason why it doesn’t maintain its structure in time indefinitely is external, rather than internal, decoherence. Speculating about their technological applications, time crystals could be employed as robust quantum memories. These may be in people’s minds this month due to the recent death of respected physicist Stephen Wiesner whose futuristic idea of quantum money relies on near-perfect quantum memories that are still unavailable today, many decades later.

Title: Quantum Computing with Circular Rydberg Atoms
Organization: Princeton University
Due to their scalability and their potential for logic gates with both multiple control and multiple target qubits, arrays of cold Rydberg atoms are among the leading platforms for quantum computing. There has been years of intense research on many fronts: circular states, cryogenic environments, ponderomotive potentials, dynamical decoupling and quantum-nondemolition detection. This work proposes a way to bring together and get the best out of all these techniques. The proposal is significant because it only involves ideas that have been experimentally demonstrated while still markedly improving the state of the art for cold atom quantum computers.

Title: Performance-Optimized Components for Quantum Technologies via Additive Manufacturing
Organizations: University of Nottingham; Jazan University; Added Scientific Ltd.
A foundational piece of equipment in many quantum technology laboratories is the optical table. These are huge, stable platforms into which lenses, mirrors and various other components can be screwed. Moving out of the lab and into the commercial world presents new opportunities for improvements. In this work, 3D printed parts were used to engineer a magneto-optical trap for cold atom experiments with enhanced stability and substantially reduced cost, size, weight and power consumption. Going forwards, the techniques that were demonstrated could benefit the stability and robustness of other quantum devices reliant on complex optics and vacuum chambers.

Title: Quantum generators of random numbers
Organization: Wrocław University of Science and Technology
This highly informative and commercially relevant article lists both the places where random numbers are useful and the times when relying on classical methods of generating pseudorandom numbers – that only approximate genuinely random numbers – has lead to cryptographic exploits. Since quantum mechanics is inherently probabilistic, a variety of companies offer quantum random number generators. After giving some background on the nature of randomness and surveying the these companies, the authors experimentally compare three quantum random number generators and a pseudorandom generator. Finally, they present a scheme to use quantum entanglement to test the randomness of a sequence in a non-destructive way.

Title: Laser-annealing Josephson junctions for yielding scaled-up superconducting quantum processors
Organization: IBM
Annealing is a heat treatment that these researchers employed to incrementally adjust the properties of a chip with numerous Josephson junctions. They used a laser, carefully controlling its power and pulse duration, so that when the device was cooled to the superconducting temperatures at which it operates, the qubit frequencies were set with a precision almost ten times greater than before the laser treatment. This improved control of the qubit frequencies will be a key enabler for superconducting quantum computers as they scale to higher numbers of qubits.


Title: Classical variational simulation of the Quantum Approximate Optimization Algorithm
Organizations: Flatiron Institute; Columbia University; École Polytechnique Fédérale de Lausanne
The Quantum Approximate Optimization Algorithm (QAOA) is a highly regarded quantum-classical algorithm that was originally designed for solving combinatorial optimization problems. In this paper, the authors classically simulate the entire algorithm using a type of neural network known as a Restricted Boltzmann Machine. Their simulation of 54 qubits used only kilobytes of storage and still agreed with the exact calculations. They could even push their simulations beyond the regime where exact results were tractable. The next step could be to explore QAOA circuits that are too large to be implemented on today’s quantum computers and thus set the stage for next generation devices.

Title: Hamiltonian simulation algorithms for near-term quantum hardware
Organizations: PhaseCraft Ltd.; University College London; University of Cambridge
In both classical and quantum computing, there are several levels of abstraction. An algorithm that was originally developed at a high level can be optimised by finding shortcuts in the lower levels. In this work, the researchers went below the level of logic gates and analysed the interactions between the qubits, reducing the circuit depth required for a certain quantum simulation from millions to hundreds. This dramatic shortening of the run-time requirements is important both for the simulation algorithms considered in the paper and for the possibility that it could be reproduced when similar analyses are performed on other quantum algorithms.

Title: Quantum-enhanced analysis of discrete stochastic processes
Organizations: Data Cybernetics; Sungkyunkwan University; University of KwaZulu-Natal; National Institute for Theoretical and Computational Sciences, South Africa
Random processes, also known as stochastic processes, can be treated in a rigorous way but doing this is often computationally expensive. Given the nature of the global financial system, the rewards for drawing correct conclusions when randomness is involved are clearly enormous. In this paper, the authors notice that n qubits can be put into a superposition of all the random outcomes that can occur in the n time steps of a discrete random process. This is an exponential speedup compared to classical Monte-Carlo methods. To illustrate their protocol in the context of financial markets, the authors successfully calculated the Delta of a European call option on a IBM quantum device consisting of five superconducting qubits. Besides finance, they suggest that the random aspects of the spread of epidemics could also be an important stochastic setting to explore in their future work.

Title: Quantum Principal Component Analysis Only Achieves an Exponential Speedup Because of Its State Preparation Assumptions
Organization: University of Washington
In 2015, Scott Aaronson wrote a Nature Physics paper called, “Read the fine print,” where the loading of the data at the start of a quantum machine learning (QML) algorithm was put under the spotlight with regards to whether it was the source of that algorithm’s speedup. In 2018, Ewin Tang began publishing a number of prominent research papers where she designed classical algorithms that were inspired by quantum algorithms and had comparable performance. In work recently accepted and published by Physical Review Letters, she focusses on the quantum algorithms for principal component analysis and for nearest-centroid clustering. By comparing on an equal footing how data get input into quantum and classical computers, she finds that these QML algorithms achieve their speedups as an artifact of their state preparation assumptions. Ultimately, this paper, especially when taken in conjunction with other insightful work, such as that on the quantum singular value transformation, greatly improves our understanding of quantum computing.

August 26, 2021