Google's Quantum Processor, named Willow, outperforms traditional computers on a cosmic timescale in terms of processing.
Google recently unveiled its newest quantum chip, named Willow, which the corporation claims can complete computations in just 5 minutes that would take the world's fastest supercomputers an unfathomable 10 septillion years. For context, the universe is only approximately 14 billion years old, a drop in the bucket compared to this timespan.
The researchers' study, published today in Nature, details the error suppression in the Willow processor and its exceptional performance. If scaled up, these findings could potentially meet the operational requirements for large-scale fault-tolerant quantum algorithms.
Quantum devices operate differently than classically programmed computers. The study highlights the reduction of errors in the Willow processor and its stellar performance, stating that if amplified, it could support the execution of large-scale fault-tolerant quantum algorithms.
Quantum technology is notorious for its sensitivity; to carry out its remarkable calculations, the system must be preserved in a quantum state. This typically requires a laboratory setting at temperatures close to absolute zero. Inside this chilly environment, the system becomes superconductive, enabling the device to surpass the limitations of classical physics.
The primary challenge (or opportunity, depending on your perspective) with quantum computers is that they are yet unable to solve problems beyond the capabilities of classical computers. That's the holy grail in quantum computing: a device capable of commercial applications that would be impractical, if not impossible, on cutting-edge classical computers.
In contrast to binary digits (bits) in classical computers, which only register a value of "0" or "1", quantum bits (or qubits) can represent both states simultaneously. Due to this property, the computer can process data quicker than its traditional counterparts. However, if too many errors emerge within the quantum system, the computation breaks down.
A significant aspect of Willow's performance is its reduced error rate as more qubits are utilized. Errors can cause quantum operations to deteriorate, but with Willow, they diminish instead of escalating with the size of the computer.
In the press release accompanying the announcement, Hartmut Neven, the founder and lead of Google Quantum AI, stated that "as we increased the size of physical qubit arrays, from a 3x3 grid to a 7x7 grid, we were able to cut the error rate in half every time, using our latest advances in quantum error correction."
"This," Neven added, "represents an exponential reduction in error rate."
The reduction in errors is known as "below threshold," and it signifies a turning point in the quest to develop future quantum computers with even fewer errors. According to a Google press release, the Willow system also demonstrated significant improvements in real-time error correction within the system—meaning that the computer was resolving errors that emerged while working on a problem. Furthermore, the qubit arrays exhibited longer lifespans than individual physical qubits within the system, suggesting that error correction was enhancing the quantum chip's overall durability.
On the random circuit sampling (RCS) benchmark, the performance of the Willow chip would far outstrip the Frontier supercomputer—the fastest classical supercomputer until last month—taking an unprecedented 10 septillion years. To frame that development, Google's Sycamore quantum computer solved a problem in 200 seconds that would have required a supercomputer roughly 10,000 years to solve, marking a significant milestone that allowed Google to claim "quantum supremacy."
In July, the quantum computing company Quantinuum revealed a 56-qubit system that surpassed the Sycamore processor on a benchmark tested in 2019, known as the linear cross entropy benchmark. Google has now pushed the envelope even further. The team employed the RCS benchmark, which measures a quantum computer's ability to outperform classical computers in calculations. Random circuit sampling may lack practical applications, but it poses a key challenge for quantum computers as they chase commercial, beyond-classical use cases.
"Even if people on Main Street don't care, it could still be very interesting," said John Preskill, the director of Caltech's Institute for Quantum Information and Matter, in a YouTube video accompanying the news. "I think quantum hardware has reached a stage now where it can advance science. We can study very complex quantum systems in a regime we've never had access to before."
"Quantum algorithms have fundamental scaling laws in their favor," Neven said. "There are similar scaling advantages for many foundational computational tasks that are essential for AI. So quantum computation will undoubtedly play a crucial role in collecting training data that's inaccessible to classical computers, training and optimizing certain learning architectures, and modeling systems where quantum effects are significant."
The Google team has now reached the third milestone in its six-step quantum roadmap towards an error-corrected quantum computer. Neven anticipates that commercial applications may be within three to five years, rather than multiple decades away. As with a qubit's actual utility, it's impossible to predict for sure—but the Willow result illustrates that tangible progress is being made.
The study on the Willow processor's error suppression suggests that with further scaling, it could potentially meet the requirements for large-scale fault-tolerant quantum algorithms in the field of physics, shaping the future of quantum technology. The reduction in errors in the Willow processor, as more qubits are utilized, is a significant achievement in quantum physics, moving us closer to the goal of error-corrected quantum computers.
