On Monday, Google’s quantum computing team released a significant paper in Nature, showcasing a key advancement in quantum error correction. The research demonstrated that increasing the number of hardware qubits assigned to a single error-corrected logical qubit led to exponential improvements in performance. In this study, a 105-qubit processor dedicated to one error-corrected qubit achieved remarkable stability, maintaining this state for an average of one hour. This progress suggests that quantum error correction can support complex algorithms that may require extended execution times.
In line with the paper’s release, Google made several announcements to highlight its ongoing commitment to quantum computing. One of these was the creation of a new fabrication facility designed specifically for the company’s superconducting processors. This new facility is a significant investment, providing better control over the manufacturing process and enabling the development of smaller test devices. The Willow processor, developed in this facility, features 105 qubits and includes design changes that enhance qubit resilience to noise, marking an important advancement in quantum hardware.
The improvements in the Willow processor are critical for reducing error rates in quantum computations. By increasing the size of individual qubits, Google made them less susceptible to noise, which is crucial for error correction. Despite acknowledging that the benchmarks used in the study were designed to highlight quantum computing’s strengths, Google is confident that quantum hardware now exceeds classical systems in benchmarks. For example, a task that would take 1025 years on a classical computer is completed in under five minutes on their new chip.
Central to the research are logical qubits, which group together multiple hardware qubits to detect and correct errors. These are essential for running complex quantum algorithms since hardware qubits frequently experience errors. Initially, adding more qubits to a logical qubit resulted in only modest improvements in error correction. However, with the new hardware improvements, increasing the number of qubits has had an exponential impact on error correction, particularly when using a specific error correction code known as the surface code.
The research found that expanding the distance between qubits in the surface code—measured as the size of the grid—significantly improved error correction. Increasing the grid size from three to five, and then seven, doubled the system’s ability to catch and correct errors with each step. This exponential improvement suggests that as hardware quality improves, scaling up the number of qubits in a logical qubit will result in even greater error suppression, a crucial development for making quantum computing more reliable.
A particularly impressive finding was that the largest logical qubit, with a distance of 15, could store quantum information for up to an hour. This stability is noteworthy considering that earlier processors were prone to frequent errors, such as those caused by cosmic rays. However, the error correction methods employed by Google were able to manage these disruptions, demonstrating the system’s robustness even in the face of external factors.
Although these advancements are significant, the research also highlights ongoing challenges in quantum computing. Despite the increased stability of logical qubits, they are still vulnerable to certain errors. Google identified two primary types of error spikes—localized temporary increases and widespread simultaneous errors. While these are rare, understanding them is crucial for further improving the reliability of quantum systems. Google also emphasized that, although error rates will never be entirely eliminated, reducing them to the point where they are practically irrelevant for most calculations is the key goal.
Google’s breakthroughs in error correction suggest that the path to practical quantum computing is becoming more achievable. With continuous improvements in hardware and error correction techniques, the company believes that there are no significant obstacles preventing the scaling of quantum systems for real-world applications. The team’s enthusiasm for these results reflects their belief that they have reached a crucial milestone in quantum computing development. The combination of hardware advances and error-correcting strategies indicates that large-scale, reliable quantum computing may soon be a reality.
