by Nadia Carlsten | on 26 OCT 2021
What if by harnessing the properties of quantum mechanics we could model and simulate the behavior of matter at its most fundamental level, down to how molecules interact? The machine that would make that possible would be transformative, changing what we know about science and how we probe nature for answers.
Quantum computers have the potential to be this machine: The scientific community has known for some time now that certain computational tasks can be solved more efficiently when qubits (quantum bits) are used to perform the calculations, and that quantum computers promise to solve some problems that are currently beyond the reach of classical computers. But many unknowns remain: How should we build such a machine so that it can handle big problems, useful problems of practical importance? How can we scale it to thousands and millions of qubits while maintaining precise control over fragile quantum states and protecting them from their environment? And what customer problems should we design it to tackle first? These are some of the big questions that motivate us at the AWS Center for Quantum Computing.
The home of AWS Quantum Technologies
In this post I am excited to announce the opening of the new home of the AWS Center for Quantum Computing, a state-of-the-art facility in Pasadena, California, where we are embarking on a journey to build a fault-tolerant quantum computer. This new building is dedicated to our quantum computing efforts, and includes office space to house our quantum research teams, and laboratories comprising the scientific equipment and specialized tools for designing and running quantum devices. Here our team of hardware engineers, quantum theorists, and software developers work side by side to tackle the many challenges of building better quantum computers. Our new facility includes everything we need to push the boundaries of quantum R&D, from making, testing, and operating quantum processors, to innovating the processes for controlling quantum computers and scaling the technologies needed to support bigger quantum devices, like cryogenic cooling systems and wiring.
From research to reality
A bold goal like building a fault-tolerant quantum computer naturally means that there will be significant scientific and engineering challenges along the way, and supporting fundamental research and making a commitment to the scientific community working on these problems is essential for accelerating progress. Our Center is located on the Caltech campus, which enables us to interact with students and faculty from leading research groups in physics and engineering just a few buildings away. We chose to partner with Caltech in part due to the university’s rich history of contributions to computing – both classical and quantum – from pioneers like Richard Feynman, whose vision 40 years ago can be credited with kick-starting the field of quantum computing, to the current technical leads of the AWS Center for Quantum Computing: Oskar Painter (John G Braun Professor of Applied Physics, Head of Quantum Hardware), and Fernando Brandao (Bren Professor of Theoretical Physics, Head of Quantum Algorithms). Through this partnership we’re also supporting the next generation of quantum scientists, by providing scholarships and training opportunities for students and young faculty members.
But our connections to the research community don’t end here. Our relationships with a diverse group of researchers help us stay at the cutting edge of quantum information sciences research. For example, several experts in quantum related fields are contributing to our efforts as Amazon Scholars and Amazon Visiting Academics, including Liang Jiang (University of Chicago), Alexey Gorshkov (University of Maryland), John Preskill (Caltech), Gil Refael (Caltech), Amir Safavi-Naeimi (Stanford), Dave Schuster (University of Chicago), and James Whitfield (Dartmouth). These experts help us innovate and overcome technical challenges even as they continue to teach and conduct research at their universities. I believe such collaborations at this early stage of the field will be critical to fully understand the potential applications and societal impact of quantum technologies.
Building a better qubit
There are many ways to physically realize a quantum computer: quantum information can, for example, be encoded in particles found in nature, such as photons or atoms, but at the AWS Center for Quantum Computing we are focusing on superconducting qubits – electrical circuit elements constructed from superconducting materials. We chose this approach partly because the ability to manufacture these qubits using well-understood microelectronic fabrication techniques makes it possible to make many qubits in a repeatable way, and gives us more control as we start scaling up the number of qubits. There is more to building a useful quantum computer than increasing the number of qubits, however. Another important metric is the computer’s clock speed, or the time required to perform quantum gate operations. Faster clock speeds means solving problems faster, and here again superconducting qubits have an edge over other modalities, as they provide very fast quantum gates.
The ultimate measure of the quality of our qubits will be the error rate, or how accurately we can perform quantum gates. Quantum devices available today are noisy and are as a result limited in the size of circuits that they can handle (a few thousands of gates is the best we can hope for with Noisy Intermediate-Scale Quantum (NISQ) devices). This in turn severely limits their computational power. There are two ways that we are approaching making better qubits at the AWS Center for Quantum Computing: the first is by improving error rates at the physical level, for example by investing in material improvements that reduce noise. The second is through innovative qubit architectures, including using Quantum Error Correction (QEC) to reduce quantum gate errors by redundantly encoding information into a protected qubit, called a logical qubit. This allows for the detection and correction of gate errors, and for the implementation of gate operations on the encoded qubits in a fault-tolerant way.
Innovating error correction
Typical QEC requires a large number of physical qubits to encode every qubit of logical information. At the AWS Center for Quantum Computing, we have been researching ways to reduce this overhead through the use of qubit architectures that allow us to implement error correction more efficiently in quantum hardware. In particular, we are optimistic about approaches that make use of linear harmonic oscillators such as Gottesman-Kitaev-Preskill (GKP) qubits and “Schrödinger cat” qubits, and recently proposed a theoretical design for a fault-tolerant quantum computer based on hardware-efficient architecture leveraging the latter.
One thing that differentiates this approach is that we take advantage of a technique called “error-biasing”. There are two types of errors that can affect quantum computation: bit-flip (flips between the 0 and 1 state due to noise) and phase-flips (the reversal of parity in the superposition of 0 and 1). In error-biasing, we use physical qubits that allow us to suppress bit-flips exponentially, while only increasing phase-flips linearly. We then combine this error-biasing with an outer repetition code consisting of a linear chain of cat qubits to detect and correct for the remaining phase-flip errors. The result is a fault-tolerant logical qubit that has a lower error rate for storing and manipulating the encoded quantum information. Not having to correct for bit-flip errors is the reason this architecture is hardware efficient and shows tremendous potential for scaling.
Building the future for our customers
The journey to an error-corrected quantum computer starts with a few logical qubits. A key milestone for our team – and the quantum computing field – will be demonstrating the breakeven point with a logical qubit, where the accuracy of the logical qubit surpasses the accuracy of the physical qubits that constitute its building blocks. Our ultimate goal is to deliver an error-corrected quantum computer that can perform reliable computations not just beyond what any classical computing technology is capable of, but at the scale needed to solve customer problems of practical importance.
Why set such an ambitious goal? The quantum algorithms that have the most potential for significant impact, for example in industries like manufacturing or pharmaceuticals, can’t be solved by simply expanding today’s quantum technologies. Pursuing breakthrough innovations rather than incremental improvements always takes longer, but I believe a bold approach that fundamentally reconsiders what makes a good qubit is the best way to deliver the ultimate computational tool: a machine that can execute algorithms requiring hundreds of thousands to billions of quantum gate operations on each qubit with at most one error over the total number of gates, a level of accuracy needed to solve the most complex computational problems that have societal and commercial value.
In talking to our AWS quantum customers over the last couple years I’ve found that those that are most excited about the potential for quantum are also realistic about the challenges of realizing the full potential of this technology, and are eager to collaborate with us to make it a reality even as they build up their own internal expertise in quantum. At the AWS Center for Quantum computing, we have assembled a fantastic team that is committed to this exciting journey toward fault-tolerant quantum computing. Stay tuned, and join us.