Sundar Pichai Introduces Willow, Quantum Computing Chip by Google
Willow is in the headlines, courtesy of Sundar Pichai, the CEO of Alphabet. He published a post on X underlining two crucial achievements of it. Firstly, Willow is a quantum computing chip by Google. Information about it has been detailed by Hartmut Neven, the Founder & Lead at Google Quantum AI. Secondly, Willow is making headlines for all the right reasons.
A brief explanation by Sundar highlighted that two problems that Willow solves are the level of errors and computation speed. The quantum computing chip by Google brings down the number of errors even if qubits are higher, and it solves standard computation in less than 5 minutes. The post on X states that this is the same problem that the modern supercomputer would take more than 10^25 years to solve.
Sundar Pichai called this an important step for them to foster implementation in different areas. This includes fusion energy, drug discovery, and battery design, to mention a few.
As per details outlined by Neven in a blog post on Google, reducing errors was a major challenge they were working on. Errors, like anywhere else, affect the speed and precision of computation. A general principle is that a higher number of qubits means more errors. This restricts the efficiency of time-based scaling. Willow solves that by bringing down the number of errors with a scaled measurement of 105 qubits.
Simply put, a larger number of qubits now only reduces the number of errors instead of increasing them. The testing phase was carried out by scaling up from 3x3 encoded qubits to 5x5 and 7x7. Error rate, according to the blog post by Neven, was slashed by half every time a test was performed.
Interestingly, this is not the only first that Google achieved. Google Quantum AI Founder said that another first achievement pertains to a compelling example of real-time error correction on a superconducting quantum system.
The result of computation is based on the RCS benchmark which is an acronym for random circuit sampling. Calling the performance on the RCS benchmark astonishing, Neven stated that the computation was done in under 5 minutes - adding that the mind-boggling number exceeds timescales in physics that are known in the current time.
However, there is a conservative assumption on which the performance of Willow has been put out for the public. The blog post on Google lists two assumptions:
Full access to secondary storage (that is hard drive); and,
No bandwidth overhead.
Also, the performance is said to have outpaced Frontier which is one of the most powerful classical supercomputers. Classical computers are expected to improve with time and the performance that is reported today will possibly have a different perspective after a couple of years.
Google is confident that Willow will make more strides as there is a broad gap between its quantum computing chip.
Promotion: Check deals for your favorite brand on Amazon.
Moving forward, Google is looking to redefine what quantum computing is by taking up a challenge: Demonstrating the first useful and beyond-classical computation on quantum chips of today’s time.
For now, the announcement has attracted the attention of Elon Musk and Vitalik Buterin. Elon Musk has commented Wow to which Sundar Pichai has responded saying that they should do a quantum cluster in space with Starship one day. Vitalik Buterin has mostly sought clarification on the largest semiprime they can factor.
A small exchange between Musk and Pichai leaves the window open for a potential association while Buterin’s response makes one think if it is a challenge or a genuine question.
Buy me a pizza if you loved this article.
Comments