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: Model-free readout-error mitigation for quantum expectation values
Organization: IBM

The paper proposes an efficient way to mitigate the readout errors that occur when the final measurements of a quantum computation are Pauli measurements. Simulations of the proposed mitigation strategy showed that more accurate expectation values could be obtained even when the readout noise was correlated. You can view the paper published on arXiv here.

Title: Quantum-accelerated multilevel Monte Carlo methods for stochastic differential equations in mathematical finance
Organizations: Phasecraft Ltd and others

The paper presents quantum-accelerated multilevel Monte Carlo methods for stochastic processes and applies them to several financial models. Theoretical analysis shows a quadratic speed-up in the precision of the computed expectation values. This is important for stochastic simulations where creating samples is costly. You can view the paper published on arXiv here.

Title: Quantum Phases of Matter on a 256-Atom Programmable Quantum Simulator
Organizations: QuEra Computing Inc. and others

Researchers have created a programmable quantum device with 256 cold atom qubits in a two-dimensional array. They expect that the size, fidelity and degree of programmability of this state-of-the-art system could be increased considerably with technical improvements. In particular, the addition of rapidly switchable local control beams would enable universal quantum computation to be performed. You can view the paper published on arXiv here Related work by the same group and a different one can be found at and at

Title: Information-theoretic bounds on quantum advantage in machine learning
Organization: AWS

The authors of this paper consider the query complexity of classical algorithms trained on data from quantum measurements. They rigorously show that for the task of making predictions with a desired average accuracy, it is comparable to optimal quantum machine learning models. However, for accurate predictions on every input, fully quantum machine learning can have an exponential advantage. The results bring the promise of both near-term and long-term quantum computing into sharper focus. You can view the paper published on arXiv here.

Title: How to Reduce the Bit-width of an Ising Model by Adding Auxiliary Spins
Organization: Waseda University

Quantum annealers, such as those made by D-wave, have the potential to provide a quantum speedup for optimization problems. However, problems of interest may require arbitrarily precise couplings and magnetic fields while the available hardware can only achieve limited precision. Researchers at Waseda University have proposed a way to reduce the bit-width of any inputs so that they can be encoded onto a device, allowing the problem-instance to be solved. You can view the paper here.

Title: A fault-tolerant continuous-variable measurement-based quantum computation architecture
Organization: AWS

The authors propose a simple yet complete architecture for fast, scalable, fault-tolerant quantum computing. They adapt an experimental set-up that has recently created some of the largest entangled states demonstrated to date – time-domain multiplexed photonic cluster states. They efficiently combine this with Gottesman-Kitaev-Preskill (GKP) encoded qubits and then test their proposal by simulating it in the presence of noise. You can view the paper published on arXiv here.

Title: Pricing Financial Derivatives with Exponential Quantum Speedup
Organizations: Quantum Mads, Santander, IQM

The Black-Scholes model is a partial differential equation describing the price of a financial option over time. The researchers map it to the Schrödinger equation and embed it in an enlarged Hilbert space. Their approach achieves an exponential speed up over classical methods. Whether it is robust to the errors that occur on NISQ devices is left for future work. You can view the paper published on arXiv here.

Title: Enhancing Combinatorial Optimization with Quantum Generative Models
Organization: Zapata Computing

Tensor network circuits are subsets of quantum circuits that may be especially well-suited to machine learning tasks. The paper explores their capability to (a) solve combinatorial optimization problems and (b) learn from other solvers, generalising and improving their attempted solutions. Promising results were achieved for financial asset allocation. You can view the paper published on arXiv here.

Title: Training variational quantum algorithms is NP-hard — even for logarithmically many qubits and free fermionic systems
Organization: Heinrich Heine University

Variational quantum algorithms (VQAs) involve a classical computer training the parameters of a quantum circuit. The goal of the classical algorithm is to find the global minimum in the training landscape that it encounters. However, the authors of this paper show that in some important settings, there are many local minima and that they are significantly worse than the global one. This adds to the question of the feasibility of training VQAs. You can view the paper published on arXiv here.

Title: Enhancing Generative Models via Quantum Correlations
Organizations: QuEra Computing Inc. and others

This work provides new insights into the design of practical quantum machine learning (ML) algorithms: start from a successful classical ML model, such as Bayesian networks or hidden Markov models, and minimally extend them by allowing local quantum measurements in bases other than the computational basis. The researchers find that these “basis-enhanced” algorithms have a higher expressive power due to the presence of quantum correlations. You can view the paper published on arXiv here.

January 25, 2021