August 26, 2020 | Written by: Jay Gambetta, Jerry Chow, and Zaira Nazario
Categorized: Quantum Computing
Share this post:
The IBM Quantum team envisions a future where devices harnessing the intricacies of atomic behavior unleash the full potential of computation unreachable due to the limits of classical computing hardware. These devices could one day transform industries from healthcare to finance to defense. We think we know the path toward that future—but it’s going to take far more than the efforts of a single company or research institution to walk that path.
Back in 2018, the National Quantum Initiative was signed into law, and today, the US Department of Energy announced that it will allocate up to $625 million in funding over the next five years to support multidisciplinary Quantum Information Science (QIS) Research Centers. We believe that these Centers closely align with our vision for the technological development of quantum computing here in the United States, both in the research that they carry out as well as the resulting national quantum ecosystem that they create. These DOE QIS Research Centers represent a crucial first effort in establishing this fundamental framework to accelerate QIS, and more specifically, quantum computing.
Quantum computers offer transformative potential in information technology, but realizing this potential requires a departure from the traditional linear model of research and development. We believe that IBM Quantum is capable of building a quantum computer with hundreds, perhaps thousands of qubits on our own—industry excels at developing and iterating on the techniques to achieve this.
However, quantum computing’s groundbreaking computational advances will require large-scale devices that compute without errors, despite the fragility of quantum information—and building such a device is beyond the limits of any one team. It will require a national effort of scientists, engineers, and the workforce supporting them, as well as expertise sharing between industry, national labs, and universities in order to build a robust quantum ecosystem. We think that realizing a fully capable, fault-tolerant quantum computer will take significant investments focused on solving three main research challenges:
- Efficient, error-corrected quantum circuits,
- Quantum interconnects, and
- Hardware-aware applications.
The main pillars of IBM Quantum Computing Research roadmap. A fully capable-fault-tolerant quantum computer relies on solutions for efficient quantum error correction, quantum interconnects, and hardware-aware applications.
First, and perhaps our grandest challenge is to demonstrate efficient, error-corrected quantum circuits. Our best theories predict this would require that we build machines with 10s of millions of qubits on a single, cooled-down chip. This is impractically high. We’ll need to co-design error correction codes with real hardware constraints and uncover new error correction strategies that can be more efficiently implemented, as even with our best effort to reduce the footprint of these qubits this is going to be a very large quantum processor. There is reason for optimism, with promising work underway indicating that we will be able to bring down the number of qubits required to implement error-correcting codes.
But even if we achieve this, we are going to need to overcome another hurdle: we’ll need to connect quantum processors with one another, just like we connect today’s computer chips inside data centers via intranets. This requires quantum interconnects that transfer the fragile quantum information stored in the processor’s qubits onto another quantum format (photons, for example) that communicate the data to another processor. Advances in this space must unite disparate technologies like superconducting qubits and fiber optics, while solving outstanding challenges in materials science and quantum communications.
Finally, building a quantum computer that can implement any circuit might be overkill. Instead, we’ll need to co-design algorithms and hardware by considering the limitations of real hardware while drafting algorithms, and simultaneously building hardware based on the needs of the algorithms we hope to run. This is only possible by espousing an agenda that aggressively pursues hardware-aware applications, advances error mitigation and compilation techniques, and develops quantum circuits in tandem.
These three main research challenges come together as focal points for our collaborations within the DOE’s National Quantum Information Science Centers, which will serve as accelerating testbeds for quantum technology across the country. Our partnership in the Centers highlight best our approach to solve these challenges as a community, in a combined effort with national labs, university labs and industries.
For example, the Co-design Center for Quantum Advantage (C2QA), aims to deliver 10x improvement each in software optimization (including building hardware-aware compilation schemes), materials and device properties, and quantum error correction to provide a 1000x improvement in computation. Working with this Center, IBM Quantum will provide quantum computing tools for prototyping, integration, testing, and benchmarking of quantum error correction codes. IBM will also help scale appropriate technologies and make them available to the Center.
Q-NEXT aims to establish efficient quantum networks for information processing with scalable hardware. IBM Quantum will work in this Center to develop crucial building blocks to accomplish that mission: technologies for quantum interconnects to link individual processors and potentially allow quantum computers to scale to billions of qubits.
Finally, the Quantum Science Center (QSC) aims to harness and control quantum coherence for new applications of QIS for sensing and scientific discovery. IBM Quantum will collaborate with this Center to design new quantum algorithms and advanced error mitigation techniques bridging the gap between near-term and fault-tolerant quantum computers, and apply them to the study of novel materials.
Furthermore, this is our chance through initiatives like these to avoid the pitfalls of previous high technology areas and cultivate a vibrant and diverse workforce to sustain this industry well into the future. The opportunities that these centers offer will have a rippling effect, driving increased demand for QIS expertise at companies and institutions across the country.
IBM offers its time-proven and vast experience in the research areas and workforce development that will be supported by these Centers. We have been involved in fundamental theoretical aspects of quantum error correction from the start. More recently, we have taken efforts toward building logical qubits through the IARPA Logical Qubits program, which takes a hardware-aware approach to developing novel error correction codes that work on realistic physical superconducting qubit hardware. Plus, IBM Quantum offers the most comprehensive and advanced quantum education program existing, worldwide, to cultivate the future quantum workforce.
We think that fault-tolerant quantum computers will become a reality, but not without long-term research and development work and a clear vision of the way forward. That’s why we’re excited about these new Quantum Information Science Research Centers and to work alongside them as they carry us into the future.
Quantum starts here