Quantum computing holds great promise as a way to address certain problems that are intractable on conventional computers. Large-scale quantum machines can also serve as a tool for scientific discovery by simulating complex quantum systems. Today, modest-scale commercial machines with dozens or even hundreds of qubits exist, but the utility of these ‘NISQ’ (Noisy Intermediate Scale Quantum) machines is limited, with no true quantum computing advantage realized to date. The realization of useful large-scale quantum computation must occur on hardware that is operated to be tolerant of faults in the computation process.
The challenge is that quantum states are very sensitive to noise because they contain information that is in some sense analog and thus inherently subject to a continuous set of possible errors (such as energy decay or dephasing of individual qubits). Quantum error correction (QEC) functions by encoding quantum states in ‘logical qubits’ comprised of multiple physical qubits in such a way that subtle measurements can be made which convert the continuous errors into discrete (and hence correctable) errors. QEC succeeds when the encoded logical qubits are preserved with lifetimes and logical gate fidelities far better than those of their constituent physical qubits. Fault tolerance is QEC that works even when the error-correcting circuit itself occasionally makes errors. Achieving fault tolerance is difficult because of the large overhead: both in the number of qubits and in the demanding performance requirements on the physical error rates. Most industrial efforts are focused on the brute-force scaling of qubit designs and architectures that were introduced over a decade ago, and are only experiencing incremental improvements today.
To advance the state of the art in quantum computing, and to speed up the development of fault-tolerant machines, new ideas and innovations are desirable at all levels of the ‘stack,’ including the hardware devices, the computing architectures and encoding schemes, and the algorithms or simulations that are developed for them. The challenging nature of this goal requires these advances to be developed synergistically, through coordinated efforts of scientists and engineers from various disciplines and backgrounds. In our Quantum Leap Challenge Institute, PRACTIQAL, we will create and test new ideas for accelerating the development of practical error-corrected and error-aware systems and train the next generation of the workforce that can translate these ideas into reality. We will achieve these goals by assembling an interlocking, multi-disciplinary team with expertise to explore the synergies between the different research areas required to address error correction needed to create the fault-tolerant systems of the future.