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: Sub-recoil clock-transition laser cooling enabling shallow optical lattice clocks
Organizations: National Institute of Standards and Technology; University of Colorado
Ever since the middle of the 20th century, the precision of atomic clocks has been improving exponentially. They have been based on cesium, which has one valence electron. Ytterbium atoms, with their two valence electrons, could be the future of atomic clocks if only better ways to cool them could be developed. This new research demonstrates the cooling of ytterbium atoms to temperatures where only a very shallow optical lattice is required to trap them. This means that the inaccuracy arising from tunneling and light shifts is reduced.
Title: Integrating quantum processor device and control optimization in a gradient-based framework
Organizations: Alibaba Quantum Laboratory; Chinese Academy of Science; Songshan Lake Materials Laboratory
Superconducting qubits come in a variety of types (eg charge, flux, phase) and they can be laid out on chips in various patterns and physically coupled together in numerous ways. Other quantum computing platforms have just as many options. While some researchers have optimized these design choices, others have concentrated on optimizing different control parameters, such as the shapes and sequences of the pulses that are sent to the devices. This latest paper explores the benefits of jointly optimizing both the device design and the control methods.
Title: An elementary quantum network of entangled optical atomic clocks
Organization: University of Oxford
In this paper, the authors improve upon a couple of previously demonstrated achievements in quantum metrology: estimating the frequency difference between two macroscopically separated atomic clocks and generating entanglement between these clocks. Performing the latter with a high fidelity and at a high rate allowed the entanglement-enhanced frequency difference of far apart clocks to be estimated for the first time. Furthermore, the precision was close to the Heisenberg limit.
Title: Ramped measurement technique for robust high-fidelity spin qubit readout
Organization: University of New South Wales
Electron spins in silicon quantum dots are one of the leading quantum computing candidates. Measurement fidelity can be above 99% when the electrons are sufficiently cold. In this paper, a new measurement technique shows similarly high fidelity but at double the temperature. Other advantages of the new method include real-time dynamic feedback and a simplified alignment scheme.
Title: Coherent quantum annealing in a programmable 2,000 qubit Ising chain
Organizations: D-Wave Systems; Saitama Medical University; University of Southern California; Tokyo Institute of Technology; Tohoku University; RIKEN; Simon Fraser University
Quantum annealing devices undergo decoherence at the onset of thermal effects from their environment. However, in this work, a quantum annealer performed a quantum simulation that gave results consistent with the annealing process being coherent. To test whether the results could also be explained by classical dynamics, several Monte Carlo simulations were performed. These could account for some, but not all, of the outputs, leading to the conclusion that the dynamics were indeed coherent. If this conclusion holds up, it would be a big step for quantum annealers since coherence is an important resource for NISQ devices.
Title: The Future of Quantum Computing with Superconducting Qubits
This paper paints a picture of how, in the absence of major advances in quantum error correction, the most feasible route to useful quantum computational advantage lies in the development of techniques like circuit knitting, error suppression, error mitigation and heuristic algorithms. Emphasis is also placed on more tightly integrating quantum and classical processors and having a low-latency, highly parallelized software stack.
Title: In situ equalization of single-atom loading in large-scale optical tweezer arrays
Organizations: Université Paris-Saclay; PASCAL; Universidad de Oviedo
Cold atom quantum computers are promising NISQ devices but one aspect where there is noticeable room for improvement is the initialisation. A small number of neutral atoms can be loaded into optical tweezers with high probability and large numbers with low probability. This new paper reports on an experimental method that assembles about 300 atoms with an unprecedented efficiency of 37%.
Title: Double-Transmon Coupler: Fast Two-Qubit Gate with No Residual Coupling for Highly Detuned Superconducting Qubits
Organization: Toshiba Corporation
Tunable couplers have become a crucial component for fast, high-fidelity, two-qubit gates in superconducting systems. They also allow one type of residual interaction to be turned off. However, for a single coupler, there remains another unwanted energy shift. This paper finds that a double-transmon coupler could eliminate this shift. According to numerical simulations, it should enable two-qubit gate fidelities of over 99.99% with a short gate time of 24 ns.
Title: Financial Index Tracking via Quantum Computing with Cardinality Constraints
Organizations: Multiverse Computing; Protiviti; Ally Financial; Donostia International Physics Center; Ikerbasque Foundation for Science
Having fewer assets in a financial portfolio can reduce management and transaction costs. This work shows how to use quantum annealing to optimize such portfolios. The researchers constructed a small set of assets that had a better risk profile than the Nasdaq-100 and S&P 500 indexes.
Title: Synergy Between Quantum Circuits and Tensor Networks: Short-cutting the Race to Practical Quantum Advantage
Organizations: Zapata Computing; Rutgers University
A quantum circuit that starts in a random initial state and has parameterised logic gates can be subject to the barren plateau phenomenon, where it is unclear how to best update the circuit’s parameters. This paper explores how, given a specific optimization task, tensor network simulations performed on classical computers can identify a promising initial state for the quantum circuit. The authors show that their method avoids barren plateaus and leads to improved performance.
Title: Covariance Matrix Preparation for Quantum Principal Component Analysis
Organizations: Universidad Autónoma de Madrid; Los Alamos National Laboratory; Quantum Science Center, Oak Ridge
Principal component analysis (PCA) is a popular tool in the data science toolkit. For certain datasets, a quantum version of PCA could have an exponential speedup over the classical version. Until this article, there was no method to encode the data into the initial quantum state. This research shows that amplitude encoding allows the covariance matrix of a dataset to be represented by probabilistically prepared quantum states. The researchers performed a numerical simulation of their method for both classical data (images of handwritten digits) and quantum data (molecular ground states).
Title: Quantum expectation-value estimation by computational basis sampling
Organizations: QunaSys Inc.; Osaka University; JST, PRESTO
Quantum circuits have to be repeated so that the measurement outcomes converge to give a precise result. This study focuses on circuits that produce a final quantum state with only a few non-negligible amplitudes. It shows how applying simple unitary transformations to the state before measuring it and combining this output with some classically efficient calculations can reduce the required number of circuit repetitions by several orders of magnitude. This approach appears to be very useful for quantum chemistry calculations and could find use cases in a variety of variational quantum algorithms.
Title: Perturbative quantum simulation
Organizations: Peking University; University of Oxford; Quantum Advantage Research; NTT Corporation; Stanford University; National University of Singapore
Perturbation theory is a method that approximates a Hamiltonian as the sum of a large, simple part and a small, complicated part. It is responsible for the majority of the quantitative predictions of quantum mechanics. In this article, the authors propose “perturbative quantum simulation,” where a NISQ processor is used to implement a perturbation theory calculation. The technique allows the processor to simulate quantum systems larger than itself. Experimental runs on the IBMQ cloud showed that the protocol is robust to noise.
Title: How Much Structure Is Needed for Huge Quantum Speedups?
Organization: University of Texas at Austin
In this survey, Scott Aaronson gives an overview of the types of tasks that can be solved exponentially faster on a quantum computer than on a classical computer. The discussion is accessible to a general scientific audience and covers both the circuit model and the oracle model. Context is given for recent, highlighted breakthroughs and possible directions for future research are presented for near-term, verifiable quantum supremacy experiments.
Title: Group-Invariant Quantum Machine Learning
Organizations: Los Alamos National Laboratory; X, Mountain View; Google Brain; University of Waterloo; Quantum Science Center, Oak Ridge
Machine learning involves a computer testing out different mathematical functions from a predefined set to find the most suitable one for a given task. If something is known about the task ahead of time (eg, whether each input is going to be an ordered list or an unordered list), then the learning process can be made more efficient. This is achieved by giving the algorithm access to only the candidate functions that are suitable for that input type. Such considerations are at the heart of this paper on quantum machine learning. By taking a systematic approach to the supervised task of binary classification, the authors recover earlier algorithms and lay the groundwork for the next steps in this field.
Title: Parallel window decoding enables scalable fault tolerant quantum computation
Organizations: Riverlane; University College London; University of Sheffield
Quantum error correction protocols proceed in rounds and for each round, a classical computer needs to perform an involved calculation. For fast quantum devices, such as superconducting quantum processors, it can be challenging for the classical computer to keep up. In this new preprint, a methodology is presented that enables the classical computer to react much faster and meet the real-time demands being placed on it.
September 30, 2022