Google’s willow chip with 105 QUBIT reaches the main quantum milestones

Google has foamy some amazing quantum computing records with its newest 105-Kube superconducting chip called Willow. This performance is not surprising, given the Google’s legacy of quantum chips for record determination, reaching Foxtail in 2017, Bristlecone in 2018 and Sycamore in 2019.

Google announced to Willow last month, and I think it is necessary to reassess the importance of this study after Jensen Huang, CEO of Nvidia recently noticed that quantum calculation is likely not to be useful for another 20 years. Anteda known, remains a lot of ground to cover guilt tolerance, which will be critical for many practical applications, but has also had a lot in Quantum in the last 12 months. Market evidence, research results (including Qub’s loyalty close to what is needed for errors tolerance) and the road maps of many quantum computing companies show that useful quantum technology is much closer than Huang believes.

Read for more how the new willow chip head in the standard of case samples. I also discuss what may be the most important part of this development for future tolerance of quantum errors, the results of applying a new corrected error code. To ensure more context, I will also share the historical perspective by Professor John Martinis, who led some of the most important jobs for previous generations of Google quantum chips, and how his work is now paid – Like he predicted – with Willow.

Improvements of Willow Hardware and Software

Willow has improved in previous generations of Google quantum chips in several ways. For starters, the use of adjustable cubes and union in Willow has provided it with much faster gates and operations that help achieve lower error levels. This speed also allows the device to be optimized or adjusted during operation. Variances in overlapping cubits can sometimes create high error rates, but tulers allow non -conforming Qubits to reconfigure and align with other Qubits to eliminate errors.

Next is the duration of quantum states. A great restriction of quantum computing has been the duration of the time that Qubits can maintain their quantum states. Willow increased that time by 5x, from 20 microseconds to 100 microseconds. This allows more complex problems to be executed.

A third advantage of Willow is that Google’s logic qub can now operate below the critical quantum error correction threshold. The QEC threshold arises from a theory developed in the 1990s, and so far has been an obstacle to efficient quantum computing. In the willow chip, however, the error rates decrease with half while the physical cubes are added to the scale. Thanks to this, as Google enhances the size of its surface code from 3×3 to 5×5 to 7×7, coded logic cubes retain their coherence for longer. Increased network size allows the most complex error patterns to be corrected, similar to more surplus in classic error correction. It also means that logic cubic could preserve their quantum states longer than the underlying physical cube.

This leads me to the only most important part of Google’s Willow announcement: Willow is the first quantum processor to demonstrate an exponential decrease in error levels as the number of QUBITS has increased. Traditionally, the addition of Qubits makes the error level increase.

Other factors needed for error -tolerant quantum calculation have also been demonstrated by Google researchers. For one thing, having a recurrent performance over several hours without degradation is necessary to execute large-scale errors-and-willow algorithms has now demonstrated that skill.

Quantum processors of comparison

Google uses case circuit samples as a continuing reference point to compare new quantum experimental processors against supercomputers who run classical algorithms. It is important to note that the sampling of the case circuit is not useful as an application in itself; Is just a threshold test. But if a system fails to exceed RCS, there is no need for further testing.

Five years ago, the Quantum Google research group claimed that its 54-sycamore’s 54-cycle superconducting Qubits (a Qub was wrong) had reached quantum superiority-it means that it exceeded classic comparable calculation. At the time, Google researchers said they were able to complete a calculation of RCS standards in 200 seconds that would theoretically need a classic super -computers to complete 10,000 years. IBM opposed the request using calculations showing that it was possible for a classic computer to achieve the same results. However, it was finally accepted by the quantum community that if Google had used all 54 QUBITs, it would have received a classic super -computer much longer than 10,000 years to equal the achievements of the Sycamore.

This year, in another quantum superiority test, Google marked the new willow chip with 105 Qubi against the same RCS standards experiment that the Sycamore Chip took place in 2019. Willow ran the RCS standard in five minutes; It is determined that today’s best classic computer would take 10 seven years to run the same landmark (this is a 1 followed by 25 zeros). In short, because Willow performs below the error correction threshold, it is able to perform casual circuit samples beyond what is possible with classic computers.

If you are not familiar with the quantum calculation, these comparisons may seem confusing at first. But they are directly attributed to the number of Qubits involved. The willow ipip has 105 qubits compared to 53 of the sycamore. Each additional Qubit results in an exponential increase in computing power, not a linear increase. The difference in the time of execution between tests in 2019 and those performed in recent months today becomes understandable in this context. Because Willow has 52 more Qubits than Sycamore, it has 2^52 (4.5 quadrilateral) more calculating states.

In addition to increasing cubes, many other improvements have been made in quantum systems since 2019. Algorithms are one billion times better due to extensive experimentation by the large community of computer scientists in the ecosystem. Plus, quantum processors have been significantly improved in various ways, including the quality of QUBITS.

Following its 2019 standards results, Google published a road map with a 10-year time limit for developing a large computer corrected with 1,000 cubic cubic errors using 1,000,000 physical cubes. As shown in the diagram above, the road map has six milestones; After her latest arrival with Willow, Google is now approaching the third point.

For another perspective on Willow’s chip, I recently discussed Google’s achievement with Prof. John Martinis, who led the Google team that drafted and tested Chip Sycamore. Prof. Martinis is currently working at a quantum beginning named Qoloab with his associates Alan Ho (another Google veteran) and Prof. Robert Mcdermott.

During that conversation, I remembered the remarks that Prof. Martinis made for a quantum computer chip still developed for a Forbes article I published nearly five years ago. “The Google plan is approximately to build a million-cub system in about 10 years, with quite low errors to make the error correction,” he said. “Then at that point you will have plenty of logical cubes corrected from errors that you can execute useful, powerful algorithms that you can now not select in a classic super computer. And maybe even in a few hundred cubits, with error Lower, it may be possible to do something special. “

These remarks are very close to describing how Google’s Willow Chip has actually played.

How long until we see quantum commercial applications?

Google currently believes it will be able to produce useful quantum trade applications in the next five years or less. Many quantum scientists believe that it will take at least a decade before quantum computers are able to handle world impact calculations in areas such as climate change, drug detection, material science and financial modeling.

Of course, Google is not the only company on this road. There is a great experiment and collaboration that is being done with logical cubes. An apparent example is Microsoft, who has done exciting work with both the blocked H-2 clogged processor and the neutral-atom processor of atom computing.

Google admits that there are many challenges left. Whereas the maximum distance of the code used in willow research was 7, to obtain the necessary degree of error tolerance would require a distance-27 cubic cubic, which would need almost 1,500 physical QUBIT to create it . For correction of quantum error, a higher distance means that an error code can handle more mistakes before it fails. A greater distance means that the code has more layers of controls and balances that can detect and repair errors before causing problems.

This is just one of the many challenges that need to be overcome to achieve errors tolerance. While some may believe that Google’s timeline is extremely optimistic, I believe the company is on the right track. In another five years, guilt tolerance will be much closer. And the useful trading quantum applications in one form or another must be quite possible.